From 4711eab6d7c5135b039e84969e4d2641fd8db33d Mon Sep 17 00:00:00 2001 From: adebowaledaniel Date: Fri, 8 Mar 2024 10:15:39 -0500 Subject: [PATCH 01/21] Add year to Covermap --- src/compare_covermaps.py | 34 ++++++++++++++++++++++++++++++++++ 1 file changed, 34 insertions(+) diff --git a/src/compare_covermaps.py b/src/compare_covermaps.py index d51a51f3..25302b9d 100644 --- a/src/compare_covermaps.py +++ b/src/compare_covermaps.py @@ -106,6 +106,8 @@ def __init__( title: str, ee_asset_str: str, resolution: int, + year=None, + collection_years=None, countries=None, probability=None, crop_labels=None, @@ -115,6 +117,8 @@ def __init__( self.ee_asset_str = ee_asset_str self.ee_asset = eval(ee_asset_str.replace("\n", "").replace(" ", "")) self.resolution = resolution + self.year = year + self.collection_years = collection_years assert (probability is None) ^ ( crop_labels is None @@ -755,18 +759,21 @@ def generate_report( .filterDate("2019-01-01", "2019-12-31")""", resolution=100, crop_labels=[40], + year=2019, ), "worldcover-v100": Covermap( "worldcover-v100", 'ee.ImageCollection("ESA/WorldCover/v100")', resolution=10, crop_labels=[40], + year=2020, ), "worldcover-v200": Covermap( "worldcover-v200", 'ee.ImageCollection("ESA/WorldCover/v200")', resolution=10, crop_labels=[40], + year=2021, ), "worldcereal-v100": Covermap( "worldcereal-v100", @@ -778,12 +785,14 @@ def generate_report( )""", resolution=10, crop_labels=[100], + year=2021, ), "glad": Covermap( "glad", 'ee.ImageCollection("users/potapovpeter/Global_cropland_2019")', resolution=30, probability=0.5, + year=2019, ), # "gfsad": Covermap( # "gfsad", @@ -797,6 +806,7 @@ def generate_report( resolution=1000, probability=100, countries=[country for country in TEST_COUNTRIES.keys() if country != "Hawaii"], + year=2017, ), "dynamicworld": Covermap( "dynamicworld", @@ -808,18 +818,21 @@ def generate_report( )""", resolution=10, crop_labels=[4], + year=2019, ), "gfsad-gcep": Covermap( "gfsad-gcep", 'ee.ImageCollection("projects/sat-io/open-datasets/GFSAD/GCEP30")', resolution=30, crop_labels=[2], + year=2015, ), "gfsad-lgrip": Covermap( "gfsad-lgrip", 'ee.ImageCollection("projects/sat-io/open-datasets/GFSAD/LGRIP30")', resolution=30, crop_labels=[2, 3], + year=2015, ), "digital-earth-africa": Covermap( "digital-earth-africa", @@ -831,6 +844,7 @@ def generate_report( resolution=10, crop_labels=[1], countries=[country for country in TEST_COUNTRIES.keys() if country != "Hawaii"], + year=2019, ), "esa-cci-africa": Covermap( "esa-cci-africa", @@ -840,6 +854,7 @@ def generate_report( resolution=20, crop_labels=[4], countries=[country for country in TEST_COUNTRIES.keys() if country != "Hawaii"], + year=2016, ), "globcover-v23": Covermap( "globcover-v23", @@ -848,6 +863,7 @@ def generate_report( )""", resolution=300, crop_labels=[11, 14, 20, 30], + year=2009, ), "globcover-v22": Covermap( "globcover-v22", @@ -856,6 +872,7 @@ def generate_report( )""", resolution=300, crop_labels=[11, 14, 20, 30], + year=2005, ), "esri-lulc": Covermap( "esri-lulc", @@ -864,6 +881,7 @@ def generate_report( ).filter(ee.Filter.date("2019-01-01", "2020-01-01"))""", resolution=10, crop_labels=[5], + year=2019, ), "nabil-etal-2021": Covermap( "nabil-etal-2021", @@ -873,6 +891,7 @@ def generate_report( resolution=30, crop_labels=[2], countries=[country for country in TEST_COUNTRIES.keys() if country != "Hawaii"], + year=2017, ), "harvest-crop-maps": Covermap( "harvest-crop-maps", @@ -880,6 +899,7 @@ def generate_report( resolution=10, probability=0.5, countries=["Togo", "Kenya", "Malawi"], + collection_years= [2019, 2019, 2020], ), "harvest-dev": Covermap( "harvest-dev", @@ -925,5 +945,19 @@ def generate_report( "Tigray2020", "Rwanda", ], + collection_years=[ + 2019, + 2020, + 2019, + 2019, + 2019, + 2020, + 2019, + 2019, + 2019, + 2020, + 2021, + 2020, + 2019,], ), } From 8e230760350c208917a0eb53c3277e1a1b788bd4 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Sat, 9 Mar 2024 04:17:39 +0000 Subject: [PATCH 02/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- src/compare_covermaps.py | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) diff --git a/src/compare_covermaps.py b/src/compare_covermaps.py index 25302b9d..00ec229b 100644 --- a/src/compare_covermaps.py +++ b/src/compare_covermaps.py @@ -825,7 +825,7 @@ def generate_report( 'ee.ImageCollection("projects/sat-io/open-datasets/GFSAD/GCEP30")', resolution=30, crop_labels=[2], - year=2015, + year=2015, ), "gfsad-lgrip": Covermap( "gfsad-lgrip", @@ -899,7 +899,7 @@ def generate_report( resolution=10, probability=0.5, countries=["Togo", "Kenya", "Malawi"], - collection_years= [2019, 2019, 2020], + collection_years=[2019, 2019, 2020], ), "harvest-dev": Covermap( "harvest-dev", @@ -958,6 +958,7 @@ def generate_report( 2020, 2021, 2020, - 2019,], + 2019, + ], ), } From d03271daf41faf178cfd69207d097e9471a92717 Mon Sep 17 00:00:00 2001 From: adebowaledaniel Date: Tue, 12 Mar 2024 10:28:01 -0400 Subject: [PATCH 03/21] template --- .../area_estimate_intercomparison.ipynb | 512 ++++++++++++++++++ 1 file changed, 512 insertions(+) create mode 100644 maps/_templates/area_estimate_intercomparison.ipynb diff --git a/maps/_templates/area_estimate_intercomparison.ipynb b/maps/_templates/area_estimate_intercomparison.ipynb new file mode 100644 index 00000000..11b4b8ff --- /dev/null +++ b/maps/_templates/area_estimate_intercomparison.ipynb @@ -0,0 +1,512 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "id": "778ff440", + "metadata": { + "id": "778ff440" + }, + "source": [ + "# Intercomparison\n", + "\n", + "**Author:**\n", + "\n", + "**Last updated:**\n", + "\n", + "**Description:** Runs intercomparison for [Country Year]\n", + "\n", + "## 1. Setup" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "fb42d13c", + "metadata": { + "id": "fb42d13c" + }, + "outputs": [], + "source": [ + "# !earthengine authenticate" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "hZ8qzSlB75kl", + "metadata": { + "id": "hZ8qzSlB75kl" + }, + "outputs": [], + "source": [ + "!git clone https://github.com/nasaharvest/crop-mask.git" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "9907f9a5", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 73 + }, + "id": "9907f9a5", + "outputId": "c9bf6e76-a345-4f21-873c-f804bc9465f4" + }, + "outputs": [], + "source": [ + "import ee\n", + "import geemap\n", + "import sys\n", + "import pandas as pd\n", + "import numpy as np\n", + "\n", + "import geopandas as gpd\n", + "from pathlib import Path\n", + "\n", + "ee.Authenticate()\n", + "ee.Initialize(project=\"bsos-geog-harvest1\")\n", + "\n", + "sys.path.append(\"../..\")\n", + "\n", + "from src.compare_covermaps import TARGETS, filter_by_bounds, generate_report, CLASS_COL, COUNTRY_COL, get_ensemble_area\n", + "from src.compare_covermaps import TEST_COUNTRIES, TEST_CODE" + ] + }, + { + "cell_type": "markdown", + "id": "c61ea4f8", + "metadata": { + "id": "c61ea4f8" + }, + "source": [ + "## 2. Read in evaluation set" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "7f75e567", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "7f75e567", + "outputId": "6bc628a4-b5b9-46f7-deae-4e19bd4cfc7e" + }, + "outputs": [], + "source": [ + "country = \"\"\n", + "\n", + "if country not in TEST_CODE:\n", + " print(f\"WARNING: {country} not found in TEST_CODE in src/compare_covermaps.py\")\n", + "if country not in TEST_COUNTRIES:\n", + " print(f\"WARNING: {country} not found in TEST_COUNTRIES in src/compare_covermaps.py\")\n", + "if country not in TEST_CODE or country not in TEST_COUNTRIES:\n", + " print(\"Please update src/compare_covermaps.py and restart the notebook.\")\n", + "else:\n", + " country_code = TEST_CODE[country]\n", + " dataset_path = \"../\" + TEST_COUNTRIES[country]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "vbVX8gFd_N3J", + "metadata": { + "id": "vbVX8gFd_N3J" + }, + "outputs": [], + "source": [ + "!dvc pull data/datasets" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "2d313baa", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "2d313baa", + "outputId": "82eca2f6-b19d-4f48-c8f8-8789b671fb89" + }, + "outputs": [], + "source": [ + "if not Path(dataset_path).exists():\n", + " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", + "else:\n", + " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", + " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", + " # use only test data because validation points used for harvest-dev map\n", + " df = df[df[\"subset\"] == \"testing\"].copy()\n", + " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", + " df[COUNTRY_COL] = country\n", + "\n", + " gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", + " gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" + ] + }, + { + "cell_type": "markdown", + "id": "31341d98", + "metadata": { + "id": "31341d98" + }, + "source": [ + "## 3. Run intercomparison" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "ImkKe6cEB4aB", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "ImkKe6cEB4aB", + "outputId": "1735fb12-fff5-4db1-e6c7-5ec405510063" + }, + "outputs": [], + "source": [ + "gdf.head()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "54c4cc0f", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "54c4cc0f", + "outputId": "9c0dde81-9fe7-4b95-ed84-cf2e843a8bbb" + }, + "outputs": [], + "source": [ + "TARGETS = {k:v for k,v in TARGETS.items()}\n", + "for k, v in TARGETS.items():\n", + " if country not in v.countries:\n", + " continue\n", + " if v.year is None:\n", + " v.year = v.collection_years[v.countries.index(country)]" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "1oQjubrHjkBi", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "1oQjubrHjkBi", + "outputId": "281b6668-e99c-420a-cdf9-dd6a6305b67c" + }, + "outputs": [], + "source": [ + "reference_year = \"\"\n", + "TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 1, reference_year, reference_year + 1]}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "98e241d2", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 142 + }, + "id": "98e241d2", + "outputId": "f4eac0c8-86d5-4145-fb60-e82795da095c" + }, + "outputs": [], + "source": [ + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", + " map_sampled = cropmap.extract_test(gdf).copy()\n", + " gdf = pd.merge(gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", + " gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "95a0f536", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 267 + }, + "id": "95a0f536", + "outputId": "194c8876-85b9-4c4b-e96f-14135eb83efc" + }, + "outputs": [], + "source": [ + "a_j = {}\n", + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", + " a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy()" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "5fJPzvOeUo9G", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "5fJPzvOeUo9G", + "outputId": "ca430174-f6b7-4165-e05d-b60bea834408" + }, + "outputs": [], + "source": [ + "# update a_j values with exported values\n", + "for cropmap in a_j.keys():\n", + " try:\n", + " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-Kenya-{cropmap}.csv')\n", + " except:\n", + " continue\n", + " crop_area = int(area_df['crop_sum'][0])\n", + " noncrop_area = int(area_df['noncrop_sum'][0])\n", + " a_j[cropmap] = np.array([noncrop_area, crop_area])" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "zyR4qCJ49Rh5", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "zyR4qCJ49Rh5", + "outputId": "687d7607-e723-4097-9d70-a3b65e149b26" + }, + "outputs": [], + "source": [ + "# Change None to nan\n", + "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "LY6Q_QtUgME_", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "LY6Q_QtUgME_", + "outputId": "38c43d9d-f9a4-4784-ad57-70c9bfc220c4" + }, + "outputs": [], + "source": [ + "from src.area_utils import compute_area_estimate, compute_area_error_matrix, compute_std_p_i\n", + "from sklearn.metrics import confusion_matrix" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "oojPqwSboiWU", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "oojPqwSboiWU", + "outputId": "bf0c96a8-477b-4e7b-9e3c-e675aba506b8" + }, + "outputs": [], + "source": [ + "# compute area estimate for each map\n", + "def compute_area_estimate(dataset, true, pred, a_j, resolution):\n", + " cm = confusion_matrix(true, pred)\n", + " total_px = a_j.sum()\n", + " w_j = a_j / total_px\n", + "\n", + " am = compute_area_error_matrix(cm, w_j)\n", + " a_i = am.sum(axis=1)\n", + " std_a_i = compute_std_p_i(w_j, am, cm)\n", + " err_a_i = 1.96 * std_a_i\n", + "\n", + " a_px = total_px * a_i\n", + " err_px = err_a_i * total_px\n", + " return pd.DataFrame(\n", + " data={\n", + " \"dataset\": dataset,\n", + " \"area_ha\": a_px[1] * (resolution**2) / (100**2),\n", + " \"err_ha\": err_px[1] * (resolution**2) / (100**2),\n", + " },\n", + " index=[0],\n", + " ).round(2)" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "ti5ZXmbyn6Mm", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 196 + }, + "id": "ti5ZXmbyn6Mm", + "outputId": "a73ecb9c-d41e-4968-e987-041cab46daea" + }, + "outputs": [], + "source": [ + "comparisons = []\n", + "area_est = []\n", + "for cropmap in TARGETS.values():\n", + " cropmap, resolution = cropmap.title, cropmap.resolution\n", + " if cropmap not in gdf.columns:\n", + " continue\n", + " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", + " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", + " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", + " comparisons.append(comparison)\n", + " area_est.append(area)\n", + "\n", + "# # Add ensemble\n", + "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", + "\n", + "# print(f\"Ensemble maps: {ensemble_maps}\")\n", + "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", + "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", + "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", + "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", + "# comparisons.append(comparison)\n", + "# area_est.append(area)\n", + "\n", + "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", + "area_est = pd.concat(area_est).set_index(['dataset'])\n", + "\n", + "results = comparisons.merge(area_est, on='dataset')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "QrAgv7pP1lcz", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "QrAgv7pP1lcz", + "outputId": "6f33c955-6ceb-4295-84ed-4aaf65c1512f" + }, + "outputs": [], + "source": [ + "results.to_csv('results.csv')" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "nAj0p7VS1_2K", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 300 + }, + "id": "nAj0p7VS1_2K", + "outputId": "78579510-3b2c-4a4e-b9e0-97cb2f9837bc" + }, + "outputs": [], + "source": [ + "results[['crop_f1','accuracy','area_ha', 'err_ha']]" + ] + }, + { + "cell_type": "markdown", + "id": "fa969373", + "metadata": { + "id": "fa969373" + }, + "source": [ + "## 4. Plot area estimate and error" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "fraQjcTMpTwp", + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 540 + }, + "id": "fraQjcTMpTwp", + "outputId": "0b4817bc-04b7-45ea-a267-0777448edc38" + }, + "outputs": [], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import matplotlib.pyplot as plt\n", + "fig, ax = plt.subplots()\n", + "# add \n", + "ax.bar(\n", + " results.index,\n", + " results[\"area_ha\"],\n", + " yerr=results[\"err_ha\"],\n", + " align=\"center\",\n", + " alpha=0.5,\n", + " ecolor=\"black\",\n", + " capsize=10,\n", + ")\n", + "ax.set_ylabel(\"Area (ha)\")\n", + "ax.set_title(\"Area of cropland\")\n", + "plt.xticks(rotation=45)\n", + "ax.yaxis.grid(True)\n", + "plt.show()" + ] + } + ], + "metadata": { + "colab": { + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.15" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} From 6b96e085982f15875169b7a787471f927f4659ad Mon Sep 17 00:00:00 2001 From: adebowaledaniel Date: Wed, 20 Mar 2024 13:30:46 -0400 Subject: [PATCH 04/21] Update notebook --- .../area_estimate_intercomparison.ipynb | 202 +++++++++++++++--- 1 file changed, 167 insertions(+), 35 deletions(-) diff --git a/maps/_templates/area_estimate_intercomparison.ipynb b/maps/_templates/area_estimate_intercomparison.ipynb index 11b4b8ff..7beebdec 100644 --- a/maps/_templates/area_estimate_intercomparison.ipynb +++ b/maps/_templates/area_estimate_intercomparison.ipynb @@ -42,6 +42,18 @@ "!git clone https://github.com/nasaharvest/crop-mask.git" ] }, + { + "cell_type": "code", + "execution_count": null, + "id": "d25d6ff7", + "metadata": {}, + "outputs": [], + "source": [ + "!pip install cartopy -qq\n", + "!pip install rasterio -qq\n", + "!pip install dvc[gs] -qq" + ] + }, { "cell_type": "code", "execution_count": null, @@ -108,7 +120,7 @@ " print(\"Please update src/compare_covermaps.py and restart the notebook.\")\n", "else:\n", " country_code = TEST_CODE[country]\n", - " dataset_path = \"../\" + TEST_COUNTRIES[country]" + " # dataset_path = \"../\" + TEST_COUNTRIES[country]" ] }, { @@ -120,7 +132,111 @@ }, "outputs": [], "source": [ - "!dvc pull data/datasets" + "# !dvc pull data/datasets" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "df7a7aaf", + "metadata": {}, + "outputs": [], + "source": [ + "ceo_set1 = \"\"\n", + "ceo_set2 = \"\"" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "c2c6cea9", + "metadata": {}, + "outputs": [], + "source": [ + "def reference_sample_agree(ceo_ref1, ceo_ref2):\n", + " ceo_ref1 = pd.read_csv(ceo_ref1)\n", + " ceo_ref2 = pd.read_csv(ceo_ref2)\n", + "\n", + " assert ceo_ref1.columns[-1] == ceo_ref2.columns[-1]\n", + "\n", + " label_question = ceo_ref1.columns[-1]\n", + "\n", + " print(f\"Number of NANs/ missing answers in set 1: {ceo_ref1[label_question].isna().sum()}\")\n", + " print(f\"Number of NANs/ missing answers in set 2: {ceo_ref2[label_question].isna().sum()}\")\n", + "\n", + " if ceo_ref1.shape[0] != ceo_ref2.shape[0]:\n", + " print(\"The number of rows in the reference sets are not equal.\")\n", + " print(\"Checking for duplictes on 'plotid'..\")\n", + " print(\n", + " \" Number of duplicated in set 1: %s\" % ceo_ref1[ceo_ref1.plotid.duplicated()].shape[0]\n", + " )\n", + " print(\n", + " \" Number of duplicated in set 2: %s\" % ceo_ref2[ceo_ref2.plotid.duplicated()].shape[0]\n", + " )\n", + " print(\"Removing duplicates and keeping the first...\")\n", + " ceo_ref1 = ceo_ref1.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", + " ceo_ref2 = ceo_ref2.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", + "\n", + " ceo_ref1.set_index(\"plotid\", inplace=True)\n", + " ceo_ref2.set_index(\"plotid\", inplace=True)\n", + " else:\n", + " print(\"The number of rows in the reference sets are equal.\")\n", + "\n", + " ceo_agree = ceo_ref1[ceo_ref1[label_question] == ceo_ref2[label_question]]\n", + "\n", + " print(\n", + " \"Number of samples that are in agreement: %d out of %d (%.2f%%)\"\n", + " % (\n", + " ceo_agree.shape[0],\n", + " ceo_ref1.shape[0],\n", + " ceo_agree.shape[0] / ceo_ref1.shape[0] * 100,\n", + " )\n", + " )\n", + " ceo_agree_geom = gpd.GeoDataFrame(\n", + " ceo_agree,\n", + " geometry=gpd.points_from_xy(ceo_agree.lon, ceo_agree.lat),\n", + " crs=\"EPSG:4326\",\n", + " )\n", + "\n", + " label_responses = ceo_agree_geom[label_question].unique()\n", + " assert len(label_responses) == 2\n", + "\n", + " for r, row in ceo_agree_geom.iterrows():\n", + "\n", + " try:\n", + " if (\n", + " row[label_question].lower() == \"crop\"\n", + " or row[label_question].lower() == \"cropland\"\n", + " or row[label_question].lower() == \"planted\"\n", + " ):\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 1\n", + " elif(\n", + " row[label_question].lower() == \"non-crop\"\n", + " or row[label_question].lower() == \"non-cropland\"\n", + " or row[label_question].lower() == \"not planted\"\n", + " ):\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 0\n", + " except IndexError:\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 255\n", + " \n", + " ceo_agree_geom = ceo_agree_geom[ceo_agree_geom[CLASS_COL] != 255]\n", + "\n", + " ceo_agree_geom[CLASS_COL] = ceo_agree_geom[CLASS_COL].astype(int)\n", + " ceo_agree_geom[COUNTRY_COL] = country\n", + " ceo_agree_geom = ceo_agree_geom[['lat','lon',CLASS_COL, COUNTRY_COL, 'geometry']]\n", + " \n", + " return ceo_agree_geom" + ] + }, + { + "cell_type": "code", + "execution_count": null, + "id": "6745bf0d", + "metadata": {}, + "outputs": [], + "source": [ + "gdf = reference_sample_agree(ceo_set1,ceo_set2)\n", + "gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" ] }, { @@ -137,18 +253,18 @@ }, "outputs": [], "source": [ - "if not Path(dataset_path).exists():\n", - " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", - "else:\n", - " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", - " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", - " # use only test data because validation points used for harvest-dev map\n", - " df = df[df[\"subset\"] == \"testing\"].copy()\n", - " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", - " df[COUNTRY_COL] = country\n", + "# if not Path(dataset_path).exists():\n", + "# print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", + "# else:\n", + "# df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", + "# df = df[(df[\"class_probability\"] != 0.5)].copy()\n", + "# # use only test data because validation points used for harvest-dev map\n", + "# df = df[df[\"subset\"] == \"testing\"].copy()\n", + "# df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", + "# df[COUNTRY_COL] = country\n", "\n", - " gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", - " gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" + "# gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", + "# gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" ] }, { @@ -288,6 +404,24 @@ " a_j[cropmap] = np.array([noncrop_area, crop_area])" ] }, + { + "cell_type": "code", + "execution_count": null, + "id": "3e853cd0", + "metadata": {}, + "outputs": [], + "source": [ + "# update a_j values with exported values\n", + "for cropmap in a_j.keys():\n", + " try:\n", + " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-Kenya-{cropmap}.csv')\n", + " except:\n", + " continue\n", + " crop_area = int(area_df['crop_sum'][0])\n", + " noncrop_area = int(area_df['noncrop_sum'][0])\n", + " a_j[cropmap] = np.array([noncrop_area, crop_area])" + ] + }, { "cell_type": "code", "execution_count": null, @@ -387,17 +521,6 @@ " comparisons.append(comparison)\n", " area_est.append(area)\n", "\n", - "# # Add ensemble\n", - "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", - "\n", - "# print(f\"Ensemble maps: {ensemble_maps}\")\n", - "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", - "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", - "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", - "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", - "# comparisons.append(comparison)\n", - "# area_est.append(area)\n", - "\n", "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", "area_est = pd.concat(area_est).set_index(['dataset'])\n", "\n", @@ -435,7 +558,7 @@ }, "outputs": [], "source": [ - "results[['crop_f1','accuracy','area_ha', 'err_ha']]" + "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']]" ] }, { @@ -462,25 +585,29 @@ }, "outputs": [], "source": [ - "import numpy as np\n", - "import matplotlib.pyplot as plt\n", - "\n", "import matplotlib.pyplot as plt\n", "fig, ax = plt.subplots()\n", - "# add \n", - "ax.bar(\n", + "\n", + "n = len(results)\n", + "colors = plt.cm.viridis(np.linspace(0, 1, n))\n", + "\n", + "ax.barh(\n", " results.index,\n", " results[\"area_ha\"],\n", - " yerr=results[\"err_ha\"],\n", + " xerr=results[\"err_ha\"],\n", " align=\"center\",\n", " alpha=0.5,\n", " ecolor=\"black\",\n", " capsize=10,\n", + " color=colors,\n", ")\n", + "\n", + "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", + " ax.text(value, i, f\"{value} ± {err}\", ha=\"center\", va=\"bottom\")\n", + "\n", "ax.set_ylabel(\"Area (ha)\")\n", "ax.set_title(\"Area of cropland\")\n", - "plt.xticks(rotation=45)\n", - "ax.yaxis.grid(True)\n", + "ax.spines[\"right\"].set_visible(False)\n", "plt.show()" ] } @@ -490,7 +617,7 @@ "provenance": [] }, "kernelspec": { - "display_name": "Python 3 (ipykernel)", + "display_name": "Python 3.9.12 ('base')", "language": "python", "name": "python3" }, @@ -504,7 +631,12 @@ "name": "python", "nbconvert_exporter": "python", "pygments_lexer": "ipython3", - "version": "3.8.15" + "version": "3.9.12" + }, + "vscode": { + "interpreter": { + "hash": "40d3a090f54c6569ab1632332b64b2c03c39dcf918b08424e98f38b5ae0af88f" + } } }, "nbformat": 4, From 6c26258c9aeaeb67cc2baf94cdf524116ee40790 Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Sat, 23 Mar 2024 17:35:08 -0400 Subject: [PATCH 05/21] Add Kenya area estimates --- maps/Kenya_2019/Kenya_area_estimate.ipynb | 3254 +++++++++++++++++++++ 1 file changed, 3254 insertions(+) create mode 100644 maps/Kenya_2019/Kenya_area_estimate.ipynb diff --git a/maps/Kenya_2019/Kenya_area_estimate.ipynb b/maps/Kenya_2019/Kenya_area_estimate.ipynb new file mode 100644 index 00000000..7b5aa483 --- /dev/null +++ b/maps/Kenya_2019/Kenya_area_estimate.ipynb @@ -0,0 +1,3254 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "778ff440" + }, + "source": [ + "# Intercomparison\n", + "\n", + "**Author:**\n", + "\n", + "**Last updated:**\n", + "\n", + "**Description:** Runs intercomparison for [Country Year]\n", + "\n", + "## 1. Setup" + ], + "id": "778ff440" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "fb42d13c" + }, + "outputs": [], + "source": [ + "# !earthengine authenticate" + ], + "id": "fb42d13c" + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "hZ8qzSlB75kl", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "80951fe8-fd11-4fd3-ce7a-f92012496b2d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Cloning into 'crop-mask'...\n", + "remote: Enumerating objects: 12074, done.\u001b[K\n", + "remote: Counting objects: 100% (1485/1485), done.\u001b[K\n", + "remote: Compressing objects: 100% (449/449), done.\u001b[K\n", + "remote: Total 12074 (delta 1102), reused 1232 (delta 1009), pack-reused 10589\u001b[K\n", + "Receiving objects: 100% (12074/12074), 125.43 MiB | 11.56 MiB/s, done.\n", + "Resolving deltas: 100% (7824/7824), done.\n", + "Updating files: 100% (208/208), done.\n" + ] + } + ], + "source": [ + "!git clone https://github.com/nasaharvest/crop-mask.git" + ], + "id": "hZ8qzSlB75kl" + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "1fe-6D3f8LTb", + "outputId": "6c6848be-2e5f-4c10-ce9c-b4dc071a2795" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "/content/crop-mask\n" + ] + } + ], + "source": [ + "%cd crop-mask/" + ], + "id": "1fe-6D3f8LTb" + }, + { + "cell_type": "code", + "source": [ + "!git checkout area-estimate-from-multi-land-cover" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "V6lTs8Z9Pt-T", + "outputId": "a9ed0471-9de0-4299-b537-069aa07a453c" + }, + "id": "V6lTs8Z9Pt-T", + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Branch 'area-estimate-from-multi-land-cover' set up to track remote branch 'area-estimate-from-multi-land-cover' from 'origin'.\n", + "Switched to a new branch 'area-estimate-from-multi-land-cover'\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "gEUyxHk9MEU2" + }, + "outputs": [], + "source": [ + "!pip install cartopy -qq\n", + "!pip install rasterio -qq\n", + "!pip install dvc[gs] -qq" + ], + "id": "gEUyxHk9MEU2" + }, + { + "cell_type": "code", + "execution_count": 5, + "metadata": { + "id": "9907f9a5", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 73 + }, + "outputId": "036bd16e-89fa-496c-84de-d90a68f3f323" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.10/dist-packages/cartopy/io/__init__.py:241: DownloadWarning: Downloading: https://naturalearth.s3.amazonaws.com/10m_cultural/ne_10m_admin_1_states_provinces.zip\n", + " warnings.warn(f'Downloading: {url}', DownloadWarning)\n" + ] + } + ], + "source": [ + "import ee\n", + "import geemap\n", + "import sys\n", + "import pandas as pd\n", + "import numpy as np\n", + "\n", + "import geopandas as gpd\n", + "from pathlib import Path\n", + "\n", + "ee.Authenticate()\n", + "ee.Initialize(project=\"bsos-geog-harvest1\")\n", + "\n", + "sys.path.append(\"../..\")\n", + "\n", + "from src.compare_covermaps import TARGETS, filter_by_bounds, generate_report, CLASS_COL, COUNTRY_COL, get_ensemble_area\n", + "from src.compare_covermaps import TEST_COUNTRIES, TEST_CODE" + ], + "id": "9907f9a5" + }, + { + "cell_type": "markdown", + "metadata": { + "id": "c61ea4f8" + }, + "source": [ + "## 2. Read in evaluation set" + ], + "id": "c61ea4f8" + }, + { + "cell_type": "code", + "execution_count": 6, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "7f75e567", + "outputId": "406be616-5fe4-4154-f113-b877ea964c93" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "country = \"Kenya\"\n", + "\n", + "if country not in TEST_CODE:\n", + " print(f\"WARNING: {country} not found in TEST_CODE in src/compare_covermaps.py\")\n", + "if country not in TEST_COUNTRIES:\n", + " print(f\"WARNING: {country} not found in TEST_COUNTRIES in src/compare_covermaps.py\")\n", + "if country not in TEST_CODE or country not in TEST_COUNTRIES:\n", + " print(\"Please update src/compare_covermaps.py and restart the notebook.\")\n", + "else:\n", + " country_code = TEST_CODE[country]\n", + " # dataset_path = \"../\" + TEST_COUNTRIES[country]" + ], + "id": "7f75e567" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "prvHkUXTOe7P", + "outputId": "a6bd565a-f50e-4d1b-cf94-ab73800059e5" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "# dataset_path = TEST_COUNTRIES[country]\n", + "dataset_path = 'data/datasets/Kenya.csv'" + ], + "id": "prvHkUXTOe7P" + }, + { + "cell_type": "code", + "source": [ + "ceo_set1 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv'\n", + "ceo_set2 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv'" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "66-YJBNxYAdF", + "outputId": "221a20ee-9808-4181-8354-d1877c544aca" + }, + "id": "66-YJBNxYAdF", + "execution_count": 7, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "def reference_sample_agree(ceo_ref1, ceo_ref2):\n", + " ceo_ref1 = pd.read_csv(ceo_ref1)\n", + " ceo_ref2 = pd.read_csv(ceo_ref2)\n", + "\n", + " assert ceo_ref1.columns[-1] == ceo_ref2.columns[-1]\n", + "\n", + " label_question = ceo_ref1.columns[-1]\n", + "\n", + " print(f\"Number of NANs/ missing answers in set 1: {ceo_ref1[label_question].isna().sum()}\")\n", + " print(f\"Number of NANs/ missing answers in set 2: {ceo_ref2[label_question].isna().sum()}\")\n", + "\n", + " if ceo_ref1.shape[0] != ceo_ref2.shape[0]:\n", + " print(\"The number of rows in the reference sets are not equal.\")\n", + " print(\"Checking for duplictes on 'plotid'..\")\n", + " print(\n", + " \" Number of duplicated in set 1: %s\" % ceo_ref1[ceo_ref1.plotid.duplicated()].shape[0]\n", + " )\n", + " print(\n", + " \" Number of duplicated in set 2: %s\" % ceo_ref2[ceo_ref2.plotid.duplicated()].shape[0]\n", + " )\n", + " print(\"Removing duplicates and keeping the first...\")\n", + " ceo_ref1 = ceo_ref1.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", + " ceo_ref2 = ceo_ref2.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", + "\n", + " ceo_ref1.set_index(\"plotid\", inplace=True)\n", + " ceo_ref2.set_index(\"plotid\", inplace=True)\n", + " else:\n", + " print(\"The number of rows in the reference sets are equal.\")\n", + "\n", + " ceo_agree = ceo_ref1[ceo_ref1[label_question] == ceo_ref2[label_question]]\n", + "\n", + " print(\n", + " \"Number of samples that are in agreement: %d out of %d (%.2f%%)\"\n", + " % (\n", + " ceo_agree.shape[0],\n", + " ceo_ref1.shape[0],\n", + " ceo_agree.shape[0] / ceo_ref1.shape[0] * 100,\n", + " )\n", + " )\n", + " ceo_agree_geom = gpd.GeoDataFrame(\n", + " ceo_agree,\n", + " geometry=gpd.points_from_xy(ceo_agree.lon, ceo_agree.lat),\n", + " crs=\"EPSG:4326\",\n", + " )\n", + "\n", + " label_responses = ceo_agree_geom[label_question].unique()\n", + " assert len(label_responses) == 2\n", + "\n", + " for r, row in ceo_agree_geom.iterrows():\n", + "\n", + " try:\n", + " if (\n", + " row[label_question].lower() == \"crop\"\n", + " or row[label_question].lower() == \"cropland\"\n", + " or row[label_question].lower() == \"planted\"\n", + " ):\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 1\n", + " elif(\n", + " row[label_question].lower() == \"non-crop\"\n", + " or row[label_question].lower() == \"non-cropland\"\n", + " or row[label_question].lower() == \"not planted\"\n", + " ):\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 0\n", + " except IndexError:\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 255\n", + "\n", + " ceo_agree_geom = ceo_agree_geom[ceo_agree_geom[CLASS_COL] != 255]\n", + "\n", + " ceo_agree_geom[CLASS_COL] = ceo_agree_geom[CLASS_COL].astype(int)\n", + " ceo_agree_geom[COUNTRY_COL] = country\n", + " ceo_agree_geom = ceo_agree_geom[['lat','lon',CLASS_COL, COUNTRY_COL, 'geometry']]\n", + "\n", + " return ceo_agree_geom" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "24QIyHfcZOeG", + "outputId": "17236484-162f-45db-a50f-e205d615f46b" + }, + "id": "24QIyHfcZOeG", + "execution_count": 8, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "gdf = reference_sample_agree(ceo_set1,ceo_set2)\n", + "gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 88 + }, + "id": "QXMdHSHVauqV", + "outputId": "a003c729-6d8f-47d8-827d-ad62206c680b" + }, + "id": "QXMdHSHVauqV", + "execution_count": 9, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Number of NANs/ missing answers in set 1: 2\n", + "Number of NANs/ missing answers in set 2: 0\n", + "The number of rows in the reference sets are equal.\n", + "Number of samples that are in agreement: 487 out of 544 (89.52%)\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "vbVX8gFd_N3J" + }, + "outputs": [], + "source": [ + "!dvc pull data/datasets" + ], + "id": "vbVX8gFd_N3J" + }, + { + "cell_type": "code", + "source": [ + "if not Path(dataset_path).exists():\n", + " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", + "else:\n", + " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", + " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", + " # use only test data because validation points used for harvest-dev map\n", + " df = df[df[\"subset\"] == \"testing\"].copy()\n", + " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", + " df[COUNTRY_COL] = country\n", + "\n", + " gdf2 = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", + " gdf2 = filter_by_bounds(country_code=country_code, gdf=gdf2)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "V8XeT-qci7VG", + "outputId": "4e9340a4-e830-4bb1-c8b6-f9bf0d741ac0" + }, + "id": "V8XeT-qci7VG", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "31341d98" + }, + "source": [ + "## 3. Run intercomparison" + ], + "id": "31341d98" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "ImkKe6cEB4aB", + "outputId": "719752af-c112-4709-e85f-f577a50d0bab" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " lat lon binary country geometry\n", + "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", + "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", + "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", + "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", + "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
latlonbinarycountrygeometry
02.40355334.9886300KenyaPOINT (34.98863 2.40355)
10.23745240.4369130KenyaPOINT (40.43691 0.23745)
24.20326135.5257850KenyaPOINT (35.52579 4.20326)
31.78613335.3450290KenyaPOINT (35.34503 1.78613)
4-2.08870741.0929161KenyaPOINT (41.09292 -2.08871)
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "gdf", + "summary": "{\n \"name\": \"gdf\",\n \"rows\": 487,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.9148026720101181,\n \"min\": -4.3996825162698165,\n \"max\": 4.509081535107014,\n \"num_unique_values\": 487,\n \"samples\": [\n -0.3086155983480622,\n 0.2319312752100853,\n -0.7040441877978902\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.9654035941659458,\n \"min\": 33.98152361411099,\n \"max\": 41.80068122597383,\n \"num_unique_values\": 487,\n \"samples\": [\n 35.89383258961963,\n 37.34840090595225,\n 35.0036305876314\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 487,\n \"samples\": [\n \"POINT (35.89383258961963 -0.3086155983480622)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 37 + } + ], + "source": [ + "gdf.head()" + ], + "id": "ImkKe6cEB4aB" + }, + { + "cell_type": "code", + "source": [ + "gdf2.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "H1JeTjxdjTNb", + "outputId": "527512fb-7321-4816-ff26-cfc167a1d162" + }, + "id": "H1JeTjxdjTNb", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " lat lon class_probability subset binary country \\\n", + "4 0.725122 34.386282 0.0 testing 0 Kenya \n", + "14 0.621939 34.466496 0.0 testing 0 Kenya \n", + "46 0.226497 34.054859 0.0 testing 0 Kenya \n", + "53 0.096834 33.951959 0.0 testing 0 Kenya \n", + "62 0.032072 33.968469 0.0 testing 0 Kenya \n", + "\n", + " geometry \n", + "4 POINT (34.38628 0.72512) \n", + "14 POINT (34.46650 0.62194) \n", + "46 POINT (34.05486 0.22650) \n", + "53 POINT (33.95196 0.09683) \n", + "62 POINT (33.96847 0.03207) " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
latlonclass_probabilitysubsetbinarycountrygeometry
40.72512234.3862820.0testing0KenyaPOINT (34.38628 0.72512)
140.62193934.4664960.0testing0KenyaPOINT (34.46650 0.62194)
460.22649734.0548590.0testing0KenyaPOINT (34.05486 0.22650)
530.09683433.9519590.0testing0KenyaPOINT (33.95196 0.09683)
620.03207233.9684690.0testing0KenyaPOINT (33.96847 0.03207)
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "gdf2", + "summary": "{\n \"name\": \"gdf2\",\n \"rows\": 764,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.9638166355465471,\n \"min\": -4.10494425,\n \"max\": 3.87641608,\n \"num_unique_values\": 763,\n \"samples\": [\n 0.39123,\n -0.86744,\n 0.476019\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.341905392717519,\n \"min\": 33.95195871,\n \"max\": 41.43769638,\n \"num_unique_values\": 764,\n \"samples\": [\n 34.37571,\n 34.65743,\n 34.23044\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.4347982228847419,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 2,\n \"samples\": [\n 1.0,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 764,\n \"samples\": [\n \"POINT (34.37571 0.39123)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 38 + } + ] + }, + { + "cell_type": "code", + "source": [ + "join_gdf = pd.concat([gdf, gdf2])\n", + "join_gdf.drop(columns=['class_probability', 'subset'])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 424 + }, + "id": "uL0EFpk5jC5M", + "outputId": "3615e7b8-39b6-4635-8104-d829cf2398f0" + }, + "id": "uL0EFpk5jC5M", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " lat lon binary country geometry\n", + "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", + "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", + "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", + "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", + "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)\n", + "... ... ... ... ... ...\n", + "8594 0.633255 34.313450 1 Kenya POINT (34.31345 0.63326)\n", + "8595 0.630631 34.314150 1 Kenya POINT (34.31415 0.63063)\n", + "8603 0.625610 34.307610 1 Kenya POINT (34.30761 0.62561)\n", + "8632 0.694220 34.369890 1 Kenya POINT (34.36989 0.69422)\n", + "8657 0.714165 34.346050 1 Kenya POINT (34.34605 0.71417)\n", + "\n", + "[1251 rows x 5 columns]" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
latlonbinarycountrygeometry
02.40355334.9886300KenyaPOINT (34.98863 2.40355)
10.23745240.4369130KenyaPOINT (40.43691 0.23745)
24.20326135.5257850KenyaPOINT (35.52579 4.20326)
31.78613335.3450290KenyaPOINT (35.34503 1.78613)
4-2.08870741.0929161KenyaPOINT (41.09292 -2.08871)
..................
85940.63325534.3134501KenyaPOINT (34.31345 0.63326)
85950.63063134.3141501KenyaPOINT (34.31415 0.63063)
86030.62561034.3076101KenyaPOINT (34.30761 0.62561)
86320.69422034.3698901KenyaPOINT (34.36989 0.69422)
86570.71416534.3460501KenyaPOINT (34.34605 0.71417)
\n", + "

1251 rows × 5 columns

\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"join_gdf\",\n \"rows\": 1251,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.411674786387463,\n \"min\": -4.3996825162698165,\n \"max\": 4.509081535107014,\n \"num_unique_values\": 1250,\n \"samples\": [\n 0.05838,\n 0.34771249,\n -3.0539655940358257\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.888586934802481,\n \"min\": 33.95195871,\n \"max\": 41.80068122597383,\n \"num_unique_values\": 1251,\n \"samples\": [\n 35.18629,\n 34.27732,\n 34.16571708\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 1251,\n \"samples\": [\n \"POINT (35.18629 -0.71086)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 40 + } + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "54c4cc0f", + "outputId": "064c956f-9a39-4d8f-f457-4b9c4af418d3" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "TARGETS = {k:v for k,v in TARGETS.items()}\n", + "for k, v in TARGETS.items():\n", + " if country not in v.countries:\n", + " continue\n", + " if v.year is None:\n", + " v.year = v.collection_years[v.countries.index(country)]" + ], + "id": "54c4cc0f" + }, + { + "cell_type": "code", + "execution_count": 11, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "1oQjubrHjkBi", + "outputId": "e8f6fd73-58ba-43ef-bd29-fdb565f30fae" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "reference_year = 2019\n", + "TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 1, reference_year, reference_year + 1]}" + ], + "id": "1oQjubrHjkBi" + }, + { + "cell_type": "code", + "execution_count": 12, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 142 + }, + "id": "98e241d2", + "outputId": "7b92af52-9833-47cd-fd9a-86de130e97d8" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[Kenya] sampling copernicus...\n", + "[Kenya] sampling worldcover-v100...\n", + "[Kenya] sampling glad...\n", + "[Kenya] sampling dynamicworld...\n", + "[Kenya] sampling digital-earth-africa...\n", + "[Kenya] sampling esri-lulc...\n", + "[Kenya] sampling harvest-crop-maps...\n" + ] + } + ], + "source": [ + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", + " map_sampled = cropmap.extract_test(gdf).copy()\n", + " gdf = pd.merge(gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", + " gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" + ], + "id": "98e241d2" + }, + { + "cell_type": "code", + "source": [ + "# for cropmap in TARGETS.values():\n", + "# if country not in cropmap.countries:\n", + "# continue\n", + "# print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", + "# map_sampled = cropmap.extract_test(join_gdf).copy()\n", + "# join_gdf = pd.merge(join_gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", + "# join_gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "l9WMBBIOjsRS", + "outputId": "15d68bc1-33da-4cab-91c8-ecebfcd5dbe6" + }, + "id": "l9WMBBIOjsRS", + "execution_count": 13, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "id": "95a0f536", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 142 + }, + "outputId": "ea96eb3c-40fe-47f1-f8db-1ebc6f39e789" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[Kenya] calculating pixel area for copernicus...\n", + "[Kenya] calculating pixel area for worldcover-v100...\n", + "[Kenya] calculating pixel area for glad...\n", + "[Kenya] calculating pixel area for dynamicworld...\n", + "[Kenya] calculating pixel area for digital-earth-africa...\n", + "[Kenya] calculating pixel area for esri-lulc...\n", + "[Kenya] calculating pixel area for harvest-crop-maps...\n" + ] + } + ], + "source": [ + "a_j = {}\n", + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", + " # a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", + " a_j[cropmap.title] = np.array([None,None])\n" + ], + "id": "95a0f536" + }, + { + "cell_type": "code", + "source": [ + "# update a_j values with exported values\n", + "for cropmap in a_j.keys():\n", + " try:\n", + " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-Kenya-{cropmap}.csv')\n", + " except:\n", + " continue\n", + " crop_area = int(area_df['crop_sum'][0])\n", + " noncrop_area = int(area_df['noncrop_sum'][0])\n", + " a_j[cropmap] = np.array([noncrop_area, crop_area])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "5fJPzvOeUo9G", + "outputId": "7b2c074f-bb39-4497-9188-277956dc0283" + }, + "id": "5fJPzvOeUo9G", + "execution_count": 15, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Change None to nan\n", + "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "zyR4qCJ49Rh5", + "outputId": "81871412-ea8a-4d19-9301-536f962cdaee" + }, + "id": "zyR4qCJ49Rh5", + "execution_count": 16, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "LY6Q_QtUgME_", + "outputId": "af2921a8-a21a-4149-a9b7-46084aec0c6b" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "from src.area_utils import compute_area_estimate, compute_area_error_matrix, compute_std_p_i\n", + "from sklearn.metrics import confusion_matrix" + ], + "id": "LY6Q_QtUgME_" + }, + { + "cell_type": "code", + "execution_count": 18, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "oojPqwSboiWU", + "outputId": "1e226f73-5983-4719-de1d-ee529a4d1c9a" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "def compute_area_estimate(dataset, true, pred, a_j, resolution):\n", + " cm = confusion_matrix(true, pred)\n", + " total_px = a_j.sum()\n", + " w_j = a_j / total_px\n", + "\n", + " am = compute_area_error_matrix(cm, w_j)\n", + " a_i = am.sum(axis=1)\n", + " std_a_i = compute_std_p_i(w_j, am, cm)\n", + " err_a_i = 1.96 * std_a_i\n", + "\n", + " a_px = total_px * a_i\n", + " err_px = err_a_i * total_px\n", + " return pd.DataFrame(\n", + " data={\n", + " \"dataset\": dataset,\n", + " \"area_ha\": a_px[1] * (resolution**2) / (100**2),\n", + " \"err_ha\": err_px[1] * (resolution**2) / (100**2),\n", + " },\n", + " index=[0],\n", + " ).round(2)" + ], + "id": "oojPqwSboiWU" + }, + { + "cell_type": "code", + "execution_count": 26, + "metadata": { + "id": "ti5ZXmbyn6Mm", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 196 + }, + "outputId": "22ed097b-76bb-444f-8d39-d41f8cb22b43" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stderr", + "text": [ + "/content/crop-mask/src/area_utils.py:385: RuntimeWarning: invalid value encountered in divide\n", + " u_j = p_jjs / p_dotjs\n", + "/content/crop-mask/src/area_utils.py:426: RuntimeWarning: invalid value encountered in divide\n", + " return p_iis / p_idots\n", + "/content/crop-mask/src/area_utils.py:454: RuntimeWarning: invalid value encountered in cast\n", + " n_i_px = ((a_j / cm.sum(axis=0) * cm).sum(axis=1)).astype(np.uint64)\n", + "/content/crop-mask/src/area_utils.py:457: RuntimeWarning: invalid value encountered in cast\n", + " n_j_px = a_j.astype(np.uint64)\n", + "/content/crop-mask/src/area_utils.py:476: RuntimeWarning: divide by zero encountered in divide\n", + " expr_3 = 1 / n_i_px**2\n" + ] + } + ], + "source": [ + "comparisons = []\n", + "area_est = []\n", + "for cropmap in TARGETS.values():\n", + " cropmap, resolution = cropmap.title, cropmap.resolution\n", + " if cropmap not in gdf.columns:\n", + " continue\n", + " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", + " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", + " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", + " comparisons.append(comparison)\n", + " area_est.append(area)\n", + "\n", + "# # Add ensemble\n", + "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", + "\n", + "# print(f\"Ensemble maps: {ensemble_maps}\")\n", + "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", + "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", + "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", + "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", + "# comparisons.append(comparison)\n", + "# area_est.append(area)\n", + "\n", + "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", + "area_est = pd.concat(area_est).set_index(['dataset'])\n", + "\n", + "results = comparisons.merge(area_est, on='dataset')" + ], + "id": "ti5ZXmbyn6Mm" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "QrAgv7pP1lcz", + "outputId": "6f33c955-6ceb-4295-84ed-4aaf65c1512f" + }, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + " \n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "results.to_csv('results.csv')" + ], + "id": "QrAgv7pP1lcz" + }, + { + "cell_type": "code", + "source": [ + "results.columns" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 124 + }, + "id": "xOO6fdt0CiG6", + "outputId": "5ba33666-7b75-4785-915c-5a2cb3a7d12f" + }, + "id": "xOO6fdt0CiG6", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Index(['country', 'crop_f1', 'std_crop_f1', 'accuracy', 'std_acc',\n", + " 'crop_recall_pa', 'std_crop_pa', 'noncrop_recall_pa', 'std_noncrop_pa',\n", + " 'crop_precision_ua', 'std_crop_ua', 'noncrop_precision_ua',\n", + " 'std_noncrop_ua', 'crop_support', 'noncrop_support', 'tn', 'fp', 'fn',\n", + " 'tp', 'tn_area', 'fp_area', 'fn_area', 'tp_area', 'area_ha', 'err_ha'],\n", + " dtype='object')" + ] + }, + "metadata": {}, + "execution_count": 21 + } + ] + }, + { + "cell_type": "code", + "execution_count": 27, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 300 + }, + "id": "nAj0p7VS1_2K", + "outputId": "b82ac7d2-451b-46e9-b76c-299f20a0178c" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", + "dataset \n", + "copernicus 0.52 0.86 0.01 0.49 0.05 \n", + "worldcover-v100 0.29 0.84 0.02 0.17 0.01 \n", + "glad 0.56 0.89 0.01 0.47 0.05 \n", + "dynamicworld NaN 0.00 NaN NaN NaN \n", + "digital-earth-africa 0.54 0.88 0.01 0.52 0.03 \n", + "esri-lulc 0.55 0.89 0.01 0.42 0.03 \n", + "harvest-crop-maps 0.47 0.92 0.01 0.47 0.05 \n", + "\n", + " crop_precision_ua std_crop_ua area_ha err_ha \n", + "dataset \n", + "copernicus 0.54 0.04 8775032.13 1674917.82 \n", + "worldcover-v100 0.91 0.04 11181466.37 1964935.17 \n", + "glad 0.69 0.04 8504157.39 1587943.81 \n", + "dynamicworld 0.00 0.00 NaN NaN \n", + "digital-earth-africa 0.56 0.04 8104337.54 1585223.55 \n", + "esri-lulc 0.76 0.04 8957165.29 1635602.01 \n", + "harvest-crop-maps 0.47 0.03 4536249.04 1366958.88 " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
crop_f1accuracystd_acccrop_recall_pastd_crop_pacrop_precision_uastd_crop_uaarea_haerr_ha
dataset
copernicus0.520.860.010.490.050.540.048775032.131674917.82
worldcover-v1000.290.840.020.170.010.910.0411181466.371964935.17
glad0.560.890.010.470.050.690.048504157.391587943.81
dynamicworldNaN0.00NaNNaNNaN0.000.00NaNNaN
digital-earth-africa0.540.880.010.520.030.560.048104337.541585223.55
esri-lulc0.550.890.010.420.030.760.048957165.291635602.01
harvest-crop-maps0.470.920.010.470.050.470.034536249.041366958.88
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI\",\n \"rows\": 7,\n \"fields\": [\n {\n \"column\": \"dataset\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 7,\n \"samples\": [\n \"copernicus\",\n \"worldcover-v100\",\n \"esri-lulc\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_f1\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.10225784403490362,\n \"min\": 0.29,\n \"max\": 0.56,\n \"num_unique_values\": 6,\n \"samples\": [\n 0.52,\n 0.29,\n 0.47\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"accuracy\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.33355944711835817,\n \"min\": 0.0,\n \"max\": 0.92,\n \"num_unique_values\": 6,\n \"samples\": [\n 0.86,\n 0.84,\n 0.92\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_acc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.00408248290463863,\n \"min\": 0.01,\n \"max\": 0.02,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.02,\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_recall_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.12832251036613437,\n \"min\": 0.17,\n \"max\": 0.52,\n \"num_unique_values\": 5,\n \"samples\": [\n 0.17,\n 0.42\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.016329931618554522,\n \"min\": 0.01,\n \"max\": 0.05,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.05,\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_precision_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.2891037052355995,\n \"min\": 0.0,\n \"max\": 0.91,\n \"num_unique_values\": 7,\n \"samples\": [\n 0.54,\n 0.91\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.014960264830861913,\n \"min\": 0.0,\n \"max\": 0.04,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.04,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"area_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 2153835.017652397,\n \"min\": 4536249.04,\n \"max\": 11181466.37,\n \"num_unique_values\": 6,\n \"samples\": [\n 8775032.13,\n 11181466.37\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"err_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 193382.44049997826,\n \"min\": 1366958.88,\n \"max\": 1964935.17,\n \"num_unique_values\": 6,\n \"samples\": [\n 1674917.82,\n 1964935.17\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 27 + } + ], + "source": [ + "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI" + ], + "id": "nAj0p7VS1_2K" + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fa969373" + }, + "source": [ + "## 4. Visualize best available map" + ], + "id": "fa969373" + }, + { + "cell_type": "code", + "source": [ + "results.dropna(inplace=True)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "qenOtnORfGTR", + "outputId": "e4430a49-6836-44be-b753-86f58ac8e387" + }, + "id": "qenOtnORfGTR", + "execution_count": 29, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 30, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "fraQjcTMpTwp", + "outputId": "df7646a2-f6f1-422d-f7ca-c430662b1b90" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "crop_proportion = round(gdf[CLASS_COL].value_counts(normalize=True)[1], 4) * 100\n", + "ax = results.sort_values(\"crop_f1\").plot(\n", + " y=[\"accuracy\", \"crop_recall_pa\", \"crop_precision_ua\", \"crop_f1\"],\n", + " xerr=\"std_crop_f1\",\n", + " kind=\"barh\",\n", + " figsize=(6, 14),\n", + " width=0.8,\n", + " title=f\"{country}: {len(gdf)} points (crop proportion: {crop_proportion}%)\",\n", + ");\n", + "\n", + "for c in ax.containers[1::2]:\n", + " ax.bar_label(c)\n", + "\n", + "for border in [\"top\", \"right\", \"bottom\", \"left\"]:\n", + " ax.spines[border].set_visible(False)\n", + "\n", + "ax.legend(bbox_to_anchor=(1, 1), reverse=True);" + ], + "id": "fraQjcTMpTwp" + }, + { + "cell_type": "code", + "source": [ + "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", + "fao_stat = fao_stat[fao_stat['Area'] == country]\n", + "fao_stat = fao_stat[fao_stat['Year Code'] == reference_year]['Value'] * 1000\n", + "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "L-nrhBekPfcp", + "outputId": "782c7067-9a3c-49c9-a6a8-608d0631e090" + }, + "id": "L-nrhBekPfcp", + "execution_count": 31, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import matplotlib.pyplot as plt\n", + "fig, ax = plt.subplots()\n", + "\n", + "n = len(results)\n", + "colors = plt.cm.viridis(np.linspace(0, 1, n))\n", + "\n", + "ax.barh(\n", + " results.index,\n", + " results[\"area_ha\"],\n", + " xerr=results[\"err_ha\"],\n", + " align=\"center\",\n", + " alpha=0.5,\n", + " ecolor=\"black\",\n", + " color= colors\n", + ")\n", + "\n", + "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", + " ax.text(value, i, f\"{value:,} ± {err:,}\", ha=\"center\", va=\"bottom\")\n", + "ax.set_ylabel(\"Area (ha)\")\n", + "ax.set_title(\"Area of cropland\")\n", + "ax.spines[\"right\"].set_visible(False)\n", + "plt.show()" + ], + "metadata": { + "id": "a0XEODxnBXW3", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 470 + }, + "outputId": "8bb01033-36f1-4ff8-f182-27c5529ac989" + }, + "id": "a0XEODxnBXW3", + "execution_count": 32, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [], + "metadata": { + "id": "gfTIBg6cwwAZ" + }, + "id": "gfTIBg6cwwAZ", + "execution_count": null, + "outputs": [] + } + ], + "metadata": { + "colab": { + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.15" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file From e80a562d9395d9cf7c9648fdd5d224d2d9b626f2 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Sat, 23 Mar 2024 21:35:44 +0000 Subject: [PATCH 06/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- maps/Kenya_2019/Kenya_area_estimate.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/maps/Kenya_2019/Kenya_area_estimate.ipynb b/maps/Kenya_2019/Kenya_area_estimate.ipynb index 7b5aa483..9caecb1c 100644 --- a/maps/Kenya_2019/Kenya_area_estimate.ipynb +++ b/maps/Kenya_2019/Kenya_area_estimate.ipynb @@ -3251,4 +3251,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From 94a30f8ad5f7c2b6b35c901fa2f36341d8258d7e Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Sat, 23 Mar 2024 17:49:56 -0400 Subject: [PATCH 07/21] Update notebook --- maps/Kenya_2019/Kenya_area_estimate.ipynb | 1351 +++------------------ 1 file changed, 139 insertions(+), 1212 deletions(-) diff --git a/maps/Kenya_2019/Kenya_area_estimate.ipynb b/maps/Kenya_2019/Kenya_area_estimate.ipynb index 9caecb1c..219e2a64 100644 --- a/maps/Kenya_2019/Kenya_area_estimate.ipynb +++ b/maps/Kenya_2019/Kenya_area_estimate.ipynb @@ -8,11 +8,11 @@ "source": [ "# Intercomparison\n", "\n", - "**Author:**\n", + "**Author:** Adebowale Adebayo\n", "\n", - "**Last updated:**\n", + "**Last updated:** March 23, 2024\n", "\n", - "**Description:** Runs intercomparison for [Country Year]\n", + "**Description:** Runs intercomparison and **area estimate** for Kenya 2019\n", "\n", "## 1. Setup" ], @@ -32,7 +32,7 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "metadata": { "id": "hZ8qzSlB75kl", "colab": { @@ -63,7 +63,7 @@ }, { "cell_type": "code", - "execution_count": 2, + "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -85,31 +85,6 @@ ], "id": "1fe-6D3f8LTb" }, - { - "cell_type": "code", - "source": [ - "!git checkout area-estimate-from-multi-land-cover" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "V6lTs8Z9Pt-T", - "outputId": "a9ed0471-9de0-4299-b537-069aa07a453c" - }, - "id": "V6lTs8Z9Pt-T", - "execution_count": 3, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Branch 'area-estimate-from-multi-land-cover' set up to track remote branch 'area-estimate-from-multi-land-cover' from 'origin'.\n", - "Switched to a new branch 'area-estimate-from-multi-land-cover'\n" - ] - } - ] - }, { "cell_type": "code", "execution_count": null, @@ -126,7 +101,7 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": null, "metadata": { "id": "9907f9a5", "colab": { @@ -177,7 +152,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -232,64 +207,10 @@ "if country not in TEST_CODE or country not in TEST_COUNTRIES:\n", " print(\"Please update src/compare_covermaps.py and restart the notebook.\")\n", "else:\n", - " country_code = TEST_CODE[country]\n", - " # dataset_path = \"../\" + TEST_COUNTRIES[country]" + " country_code = TEST_CODE[country]" ], "id": "7f75e567" }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "prvHkUXTOe7P", - "outputId": "a6bd565a-f50e-4d1b-cf94-ab73800059e5" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ], - "source": [ - "# dataset_path = TEST_COUNTRIES[country]\n", - "dataset_path = 'data/datasets/Kenya.csv'" - ], - "id": "prvHkUXTOe7P" - }, { "cell_type": "code", "source": [ @@ -305,7 +226,7 @@ "outputId": "221a20ee-9808-4181-8354-d1877c544aca" }, "id": "66-YJBNxYAdF", - "execution_count": 7, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -429,203 +350,7 @@ "outputId": "17236484-162f-45db-a50f-e205d615f46b" }, "id": "24QIyHfcZOeG", - "execution_count": 8, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "source": [ - "gdf = reference_sample_agree(ceo_set1,ceo_set2)\n", - "gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 88 - }, - "id": "QXMdHSHVauqV", - "outputId": "a003c729-6d8f-47d8-827d-ad62206c680b" - }, - "id": "QXMdHSHVauqV", - "execution_count": 9, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Number of NANs/ missing answers in set 1: 2\n", - "Number of NANs/ missing answers in set 2: 0\n", - "The number of rows in the reference sets are equal.\n", - "Number of samples that are in agreement: 487 out of 544 (89.52%)\n" - ] - } - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "id": "vbVX8gFd_N3J" - }, - "outputs": [], - "source": [ - "!dvc pull data/datasets" - ], - "id": "vbVX8gFd_N3J" - }, - { - "cell_type": "code", - "source": [ - "if not Path(dataset_path).exists():\n", - " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", - "else:\n", - " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", - " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", - " # use only test data because validation points used for harvest-dev map\n", - " df = df[df[\"subset\"] == \"testing\"].copy()\n", - " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", - " df[COUNTRY_COL] = country\n", - "\n", - " gdf2 = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", - " gdf2 = filter_by_bounds(country_code=country_code, gdf=gdf2)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "V8XeT-qci7VG", - "outputId": "4e9340a4-e830-4bb1-c8b6-f9bf0d741ac0" - }, - "id": "V8XeT-qci7VG", - "execution_count": null, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "markdown", - "metadata": { - "id": "31341d98" - }, - "source": [ - "## 3. Run intercomparison" - ], - "id": "31341d98" - }, - { - "cell_type": "code", "execution_count": null, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 206 - }, - "id": "ImkKe6cEB4aB", - "outputId": "719752af-c112-4709-e85f-f577a50d0bab" - }, "outputs": [ { "output_type": "display_data", @@ -651,711 +376,102 @@ " .geemap-colab {\n", " background-color: var(--colab-primary-surface-color, white);\n", " }\n", - "\n", - " .geemap-colab .jupyter-button {\n", - " --jp-layout-color3: var(--colab-primary-surface-color, white);\n", - " }\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "execute_result", - "data": { - "text/plain": [ - " lat lon binary country geometry\n", - "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", - "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", - "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", - "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", - "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)" - ], - "text/html": [ - "\n", - "
\n", - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
latlonbinarycountrygeometry
02.40355334.9886300KenyaPOINT (34.98863 2.40355)
10.23745240.4369130KenyaPOINT (40.43691 0.23745)
24.20326135.5257850KenyaPOINT (35.52579 4.20326)
31.78613335.3450290KenyaPOINT (35.34503 1.78613)
4-2.08870741.0929161KenyaPOINT (41.09292 -2.08871)
\n", - "
\n", - "
\n", - "\n", - "
\n", - " \n", - "\n", - " \n", - "\n", - " \n", - "
\n", - "\n", - "\n", - "
\n", - " \n", - "\n", - "\n", - "\n", - " \n", - "
\n", - "
\n", - "
\n" - ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "variable_name": "gdf", - "summary": "{\n \"name\": \"gdf\",\n \"rows\": 487,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.9148026720101181,\n \"min\": -4.3996825162698165,\n \"max\": 4.509081535107014,\n \"num_unique_values\": 487,\n \"samples\": [\n -0.3086155983480622,\n 0.2319312752100853,\n -0.7040441877978902\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.9654035941659458,\n \"min\": 33.98152361411099,\n \"max\": 41.80068122597383,\n \"num_unique_values\": 487,\n \"samples\": [\n 35.89383258961963,\n 37.34840090595225,\n 35.0036305876314\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 487,\n \"samples\": [\n \"POINT (35.89383258961963 -0.3086155983480622)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } - }, - "metadata": {}, - "execution_count": 37 - } - ], - "source": [ - "gdf.head()" - ], - "id": "ImkKe6cEB4aB" - }, - { - "cell_type": "code", - "source": [ - "gdf2.head()" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 206 - }, - "id": "H1JeTjxdjTNb", - "outputId": "527512fb-7321-4816-ff26-cfc167a1d162" - }, - "id": "H1JeTjxdjTNb", - "execution_count": null, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "execute_result", - "data": { - "text/plain": [ - " lat lon class_probability subset binary country \\\n", - "4 0.725122 34.386282 0.0 testing 0 Kenya \n", - "14 0.621939 34.466496 0.0 testing 0 Kenya \n", - "46 0.226497 34.054859 0.0 testing 0 Kenya \n", - "53 0.096834 33.951959 0.0 testing 0 Kenya \n", - "62 0.032072 33.968469 0.0 testing 0 Kenya \n", - "\n", - " geometry \n", - "4 POINT (34.38628 0.72512) \n", - "14 POINT (34.46650 0.62194) \n", - "46 POINT (34.05486 0.22650) \n", - "53 POINT (33.95196 0.09683) \n", - "62 POINT (33.96847 0.03207) " - ], - "text/html": [ - "\n", - "
\n", - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
latlonclass_probabilitysubsetbinarycountrygeometry
40.72512234.3862820.0testing0KenyaPOINT (34.38628 0.72512)
140.62193934.4664960.0testing0KenyaPOINT (34.46650 0.62194)
460.22649734.0548590.0testing0KenyaPOINT (34.05486 0.22650)
530.09683433.9519590.0testing0KenyaPOINT (33.95196 0.09683)
620.03207233.9684690.0testing0KenyaPOINT (33.96847 0.03207)
\n", - "
\n", - "
\n", - "\n", - "
\n", - " \n", - "\n", - " \n", - "\n", - " \n", - "
\n", - "\n", - "\n", - "
\n", - " \n", - "\n", - "\n", - "\n", - " \n", - "
\n", - "
\n", - "
\n" - ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "variable_name": "gdf2", - "summary": "{\n \"name\": \"gdf2\",\n \"rows\": 764,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.9638166355465471,\n \"min\": -4.10494425,\n \"max\": 3.87641608,\n \"num_unique_values\": 763,\n \"samples\": [\n 0.39123,\n -0.86744,\n 0.476019\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.341905392717519,\n \"min\": 33.95195871,\n \"max\": 41.43769638,\n \"num_unique_values\": 764,\n \"samples\": [\n 34.37571,\n 34.65743,\n 34.23044\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.4347982228847419,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 2,\n \"samples\": [\n 1.0,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 764,\n \"samples\": [\n \"POINT (34.37571 0.39123)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } + "\n", + " .geemap-colab .jupyter-button {\n", + " --jp-layout-color3: var(--colab-primary-surface-color, white);\n", + " }\n", + " \n", + " " + ] }, - "metadata": {}, - "execution_count": 38 + "metadata": {} } ] }, { "cell_type": "code", "source": [ - "join_gdf = pd.concat([gdf, gdf2])\n", - "join_gdf.drop(columns=['class_probability', 'subset'])" + "gdf = reference_sample_agree(ceo_set1,ceo_set2)\n", + "gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/", - "height": 424 + "height": 88 + }, + "id": "QXMdHSHVauqV", + "outputId": "a003c729-6d8f-47d8-827d-ad62206c680b" + }, + "id": "QXMdHSHVauqV", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} }, - "id": "uL0EFpk5jC5M", - "outputId": "3615e7b8-39b6-4635-8104-d829cf2398f0" + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Number of NANs/ missing answers in set 1: 2\n", + "Number of NANs/ missing answers in set 2: 0\n", + "The number of rows in the reference sets are equal.\n", + "Number of samples that are in agreement: 487 out of 544 (89.52%)\n" + ] + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "31341d98" }, - "id": "uL0EFpk5jC5M", + "source": [ + "## 3. Run intercomparison and area estimation" + ], + "id": "31341d98" + }, + { + "cell_type": "code", "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "ImkKe6cEB4aB", + "outputId": "719752af-c112-4709-e85f-f577a50d0bab" + }, "outputs": [ { "output_type": "display_data", @@ -1395,24 +511,16 @@ "output_type": "execute_result", "data": { "text/plain": [ - " lat lon binary country geometry\n", - "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", - "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", - "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", - "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", - "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)\n", - "... ... ... ... ... ...\n", - "8594 0.633255 34.313450 1 Kenya POINT (34.31345 0.63326)\n", - "8595 0.630631 34.314150 1 Kenya POINT (34.31415 0.63063)\n", - "8603 0.625610 34.307610 1 Kenya POINT (34.30761 0.62561)\n", - "8632 0.694220 34.369890 1 Kenya POINT (34.36989 0.69422)\n", - "8657 0.714165 34.346050 1 Kenya POINT (34.34605 0.71417)\n", - "\n", - "[1251 rows x 5 columns]" + " lat lon binary country geometry\n", + "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", + "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", + "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", + "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", + "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)" ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "execution_count": 14, + "execution_count": null, "metadata": { "id": "95a0f536", "colab": { @@ -2055,8 +1061,7 @@ " if country not in cropmap.countries:\n", " continue\n", " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", - " # a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", - " a_j[cropmap.title] = np.array([None,None])\n" + " a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy()" ], "id": "95a0f536" }, @@ -2082,7 +1087,7 @@ "outputId": "7b2c074f-bb39-4497-9188-277956dc0283" }, "id": "5fJPzvOeUo9G", - "execution_count": 15, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -2135,7 +1140,7 @@ "outputId": "81871412-ea8a-4d19-9301-536f962cdaee" }, "id": "zyR4qCJ49Rh5", - "execution_count": 16, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -2175,7 +1180,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -2228,7 +1233,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -2299,7 +1304,7 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": null, "metadata": { "id": "ti5ZXmbyn6Mm", "colab": { @@ -2373,17 +1378,6 @@ " comparisons.append(comparison)\n", " area_est.append(area)\n", "\n", - "# # Add ensemble\n", - "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", - "\n", - "# print(f\"Ensemble maps: {ensemble_maps}\")\n", - "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", - "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", - "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", - "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", - "# comparisons.append(comparison)\n", - "# area_est.append(area)\n", - "\n", "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", "area_est = pd.concat(area_est).set_index(['dataset'])\n", "\n", @@ -2445,74 +1439,7 @@ }, { "cell_type": "code", - "source": [ - "results.columns" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 124 - }, - "id": "xOO6fdt0CiG6", - "outputId": "5ba33666-7b75-4785-915c-5a2cb3a7d12f" - }, - "id": "xOO6fdt0CiG6", "execution_count": null, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "execute_result", - "data": { - "text/plain": [ - "Index(['country', 'crop_f1', 'std_crop_f1', 'accuracy', 'std_acc',\n", - " 'crop_recall_pa', 'std_crop_pa', 'noncrop_recall_pa', 'std_noncrop_pa',\n", - " 'crop_precision_ua', 'std_crop_ua', 'noncrop_precision_ua',\n", - " 'std_noncrop_ua', 'crop_support', 'noncrop_support', 'tn', 'fp', 'fn',\n", - " 'tp', 'tn_area', 'fp_area', 'fn_area', 'tp_area', 'area_ha', 'err_ha'],\n", - " dtype='object')" - ] - }, - "metadata": {}, - "execution_count": 21 - } - ] - }, - { - "cell_type": "code", - "execution_count": 27, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -2931,20 +1858,10 @@ } ], "source": [ - "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI" + "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']]" ], "id": "nAj0p7VS1_2K" }, - { - "cell_type": "markdown", - "metadata": { - "id": "fa969373" - }, - "source": [ - "## 4. Visualize best available map" - ], - "id": "fa969373" - }, { "cell_type": "code", "source": [ @@ -2959,7 +1876,7 @@ "outputId": "e4430a49-6836-44be-b753-86f58ac8e387" }, "id": "qenOtnORfGTR", - "execution_count": 29, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -2999,7 +1916,7 @@ }, { "cell_type": "code", - "execution_count": 30, + "execution_count": null, "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -3075,10 +1992,20 @@ ], "id": "fraQjcTMpTwp" }, + { + "cell_type": "markdown", + "source": [ + "## 4. Cropland Area estimates comparsion (FAOSTAT-cropland area included)" + ], + "metadata": { + "id": "ya4gJVqRQYPh" + }, + "id": "ya4gJVqRQYPh" + }, { "cell_type": "code", "source": [ - "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", + "fao_stat = pd.read_csv(\"./data/ref_samples_area/FAOSTAT_data_en_3-13-2024.csv\")\n", "fao_stat = fao_stat[fao_stat['Area'] == country]\n", "fao_stat = fao_stat[fao_stat['Year Code'] == reference_year]['Value'] * 1000\n", "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" @@ -3092,7 +2019,7 @@ "outputId": "782c7067-9a3c-49c9-a6a8-608d0631e090" }, "id": "L-nrhBekPfcp", - "execution_count": 31, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -3168,7 +2095,7 @@ "outputId": "8bb01033-36f1-4ff8-f182-27c5529ac989" }, "id": "a0XEODxnBXW3", - "execution_count": 32, + "execution_count": null, "outputs": [ { "output_type": "display_data", @@ -3251,4 +2178,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file From a1a6e2e08c4a3eb04425f8d7a8d056b52db6ea83 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Sat, 23 Mar 2024 21:50:15 +0000 Subject: [PATCH 08/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- maps/Kenya_2019/Kenya_area_estimate.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/maps/Kenya_2019/Kenya_area_estimate.ipynb b/maps/Kenya_2019/Kenya_area_estimate.ipynb index 219e2a64..a2e04969 100644 --- a/maps/Kenya_2019/Kenya_area_estimate.ipynb +++ b/maps/Kenya_2019/Kenya_area_estimate.ipynb @@ -2178,4 +2178,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From 7f4e90a7abf64564f96a7793692924b8edddeca9 Mon Sep 17 00:00:00 2001 From: adebowaledaniel Date: Sat, 23 Mar 2024 17:58:13 -0400 Subject: [PATCH 09/21] Add ref samples --- .../FAOSTAT_data_en_3-13-2024.csv | 97 ++++ ...le-2019---Set-1-sample-data-2024-03-14.csv | 545 ++++++++++++++++++ ...le-2019---Set-2-sample-data-2024-03-14.csv | 545 ++++++++++++++++++ 3 files changed, 1187 insertions(+) create mode 100644 data/ref_samples_area/FAOSTAT_data_en_3-13-2024.csv create mode 100644 data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv create mode 100644 data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv diff --git a/data/ref_samples_area/FAOSTAT_data_en_3-13-2024.csv b/data/ref_samples_area/FAOSTAT_data_en_3-13-2024.csv new file mode 100644 index 00000000..8ad66929 --- /dev/null +++ b/data/ref_samples_area/FAOSTAT_data_en_3-13-2024.csv @@ -0,0 +1,97 @@ +Domain Code,Domain,Area Code (M49),Area,Element Code,Element,Item Code,Item,Year Code,Year,Unit,Value,Flag,Flag Description,Note +"RL","Land Use","108","Burundi","5110","Area","6620","Cropland","2018","2018","1000 ha","1539","I","Imputed value","" +"RL","Land Use","108","Burundi","5110","Area","6620","Cropland","2019","2019","1000 ha","1566","I","Imputed value","" +"RL","Land Use","108","Burundi","5110","Area","6620","Cropland","2020","2020","1000 ha","1593","I","Imputed value","" +"RL","Land Use","108","Burundi","5110","Area","6620","Cropland","2021","2021","1000 ha","1620","I","Imputed value","" +"RL","Land Use","174","Comoros","5110","Area","6620","Cropland","2018","2018","1000 ha","118","I","Imputed value","" +"RL","Land Use","174","Comoros","5110","Area","6620","Cropland","2019","2019","1000 ha","118","I","Imputed value","" +"RL","Land Use","174","Comoros","5110","Area","6620","Cropland","2020","2020","1000 ha","118","I","Imputed value","" +"RL","Land Use","174","Comoros","5110","Area","6620","Cropland","2021","2021","1000 ha","118","I","Imputed value","" +"RL","Land Use","262","Djibouti","5110","Area","6620","Cropland","2018","2018","1000 ha","2.8","I","Imputed value","" +"RL","Land Use","262","Djibouti","5110","Area","6620","Cropland","2019","2019","1000 ha","2.9","I","Imputed value","" +"RL","Land Use","262","Djibouti","5110","Area","6620","Cropland","2020","2020","1000 ha","3.9","I","Imputed value","" +"RL","Land Use","262","Djibouti","5110","Area","6620","Cropland","2021","2021","1000 ha","3.9","I","Imputed value","" +"RL","Land Use","232","Eritrea","5110","Area","6620","Cropland","2018","2018","1000 ha","692","I","Imputed value","" +"RL","Land Use","232","Eritrea","5110","Area","6620","Cropland","2019","2019","1000 ha","692","I","Imputed value","" +"RL","Land Use","232","Eritrea","5110","Area","6620","Cropland","2020","2020","1000 ha","692","I","Imputed value","" +"RL","Land Use","232","Eritrea","5110","Area","6620","Cropland","2021","2021","1000 ha","692","I","Imputed value","" +"RL","Land Use","231","Ethiopia","5110","Area","6620","Cropland","2018","2018","1000 ha","17903","A","Official figure","" +"RL","Land Use","231","Ethiopia","5110","Area","6620","Cropland","2019","2019","1000 ha","18190","I","Imputed value","" +"RL","Land Use","231","Ethiopia","5110","Area","6620","Cropland","2020","2020","1000 ha","18476.1","A","Official figure","" +"RL","Land Use","231","Ethiopia","5110","Area","6620","Cropland","2021","2021","1000 ha","18595","I","Imputed value","" +"RL","Land Use","404","Kenya","5110","Area","6620","Cropland","2018","2018","1000 ha","6399","I","Imputed value","" +"RL","Land Use","404","Kenya","5110","Area","6620","Cropland","2019","2019","1000 ha","6410","I","Imputed value","" +"RL","Land Use","404","Kenya","5110","Area","6620","Cropland","2020","2020","1000 ha","6410","I","Imputed value","" +"RL","Land Use","404","Kenya","5110","Area","6620","Cropland","2021","2021","1000 ha","6410","I","Imputed value","" +"RL","Land Use","450","Madagascar","5110","Area","6620","Cropland","2018","2018","1000 ha","3600","I","Imputed value","" +"RL","Land Use","450","Madagascar","5110","Area","6620","Cropland","2019","2019","1000 ha","3600","I","Imputed value","" +"RL","Land Use","450","Madagascar","5110","Area","6620","Cropland","2020","2020","1000 ha","3600","I","Imputed value","" +"RL","Land Use","450","Madagascar","5110","Area","6620","Cropland","2021","2021","1000 ha","3600","I","Imputed value","" +"RL","Land Use","454","Malawi","5110","Area","6620","Cropland","2018","2018","1000 ha","4023","I","Imputed value","" +"RL","Land Use","454","Malawi","5110","Area","6620","Cropland","2019","2019","1000 ha","4078","I","Imputed value","" +"RL","Land Use","454","Malawi","5110","Area","6620","Cropland","2020","2020","1000 ha","4134","I","Imputed value","" +"RL","Land Use","454","Malawi","5110","Area","6620","Cropland","2021","2021","1000 ha","4200","I","Imputed value","" +"RL","Land Use","480","Mauritius","5110","Area","6620","Cropland","2018","2018","1000 ha","79","I","Imputed value","" +"RL","Land Use","480","Mauritius","5110","Area","6620","Cropland","2019","2019","1000 ha","79","I","Imputed value","" +"RL","Land Use","480","Mauritius","5110","Area","6620","Cropland","2020","2020","1000 ha","79","I","Imputed value","" +"RL","Land Use","480","Mauritius","5110","Area","6620","Cropland","2021","2021","1000 ha","79","I","Imputed value","" +"RL","Land Use","175","Mayotte","5110","Area","6620","Cropland","2018","2018","1000 ha","19.98","A","Official figure","" +"RL","Land Use","175","Mayotte","5110","Area","6620","Cropland","2019","2019","1000 ha","19.98","A","Official figure","" +"RL","Land Use","175","Mayotte","5110","Area","6620","Cropland","2020","2020","1000 ha","19.98","A","Official figure","" +"RL","Land Use","175","Mayotte","5110","Area","6620","Cropland","2021","2021","1000 ha","19.98","I","Imputed value","" +"RL","Land Use","508","Mozambique","5110","Area","6620","Cropland","2018","2018","1000 ha","5950","I","Imputed value","" +"RL","Land Use","508","Mozambique","5110","Area","6620","Cropland","2019","2019","1000 ha","5950","I","Imputed value","" +"RL","Land Use","508","Mozambique","5110","Area","6620","Cropland","2020","2020","1000 ha","5950","I","Imputed value","" +"RL","Land Use","508","Mozambique","5110","Area","6620","Cropland","2021","2021","1000 ha","5950","I","Imputed value","" +"RL","Land Use","516","Namibia","5110","Area","6620","Cropland","2018","2018","1000 ha","811","I","Imputed value","" +"RL","Land Use","516","Namibia","5110","Area","6620","Cropland","2019","2019","1000 ha","811","I","Imputed value","" +"RL","Land Use","516","Namibia","5110","Area","6620","Cropland","2020","2020","1000 ha","811","I","Imputed value","" +"RL","Land Use","516","Namibia","5110","Area","6620","Cropland","2021","2021","1000 ha","812","I","Imputed value","" +"RL","Land Use","638","Réunion","5110","Area","6620","Cropland","2018","2018","1000 ha","37.28","A","Official figure","" +"RL","Land Use","638","Réunion","5110","Area","6620","Cropland","2019","2019","1000 ha","37.14","A","Official figure","" +"RL","Land Use","638","Réunion","5110","Area","6620","Cropland","2020","2020","1000 ha","36.82","A","Official figure","" +"RL","Land Use","638","Réunion","5110","Area","6620","Cropland","2021","2021","1000 ha","36.82","I","Imputed value","" +"RL","Land Use","646","Rwanda","5110","Area","6620","Cropland","2018","2018","1000 ha","1545","I","Imputed value","" +"RL","Land Use","646","Rwanda","5110","Area","6620","Cropland","2019","2019","1000 ha","1502","I","Imputed value","" +"RL","Land Use","646","Rwanda","5110","Area","6620","Cropland","2020","2020","1000 ha","1560.6","I","Imputed value","" +"RL","Land Use","646","Rwanda","5110","Area","6620","Cropland","2021","2021","1000 ha","1618.4","I","Imputed value","" +"RL","Land Use","686","Senegal","5110","Area","6620","Cropland","2018","2018","1000 ha","3758","I","Imputed value","" +"RL","Land Use","686","Senegal","5110","Area","6620","Cropland","2019","2019","1000 ha","3795","I","Imputed value","" +"RL","Land Use","686","Senegal","5110","Area","6620","Cropland","2020","2020","1000 ha","3830","I","Imputed value","" +"RL","Land Use","686","Senegal","5110","Area","6620","Cropland","2021","2021","1000 ha","3911","I","Imputed value","" +"RL","Land Use","690","Seychelles","5110","Area","6620","Cropland","2018","2018","1000 ha","1.55","I","Imputed value","" +"RL","Land Use","690","Seychelles","5110","Area","6620","Cropland","2019","2019","1000 ha","1.55","I","Imputed value","" +"RL","Land Use","690","Seychelles","5110","Area","6620","Cropland","2020","2020","1000 ha","1.55","I","Imputed value","" +"RL","Land Use","690","Seychelles","5110","Area","6620","Cropland","2021","2021","1000 ha","1.55","I","Imputed value","" +"RL","Land Use","706","Somalia","5110","Area","6620","Cropland","2018","2018","1000 ha","1126","I","Imputed value","" +"RL","Land Use","706","Somalia","5110","Area","6620","Cropland","2019","2019","1000 ha","1127","I","Imputed value","" +"RL","Land Use","706","Somalia","5110","Area","6620","Cropland","2020","2020","1000 ha","1128","I","Imputed value","" +"RL","Land Use","706","Somalia","5110","Area","6620","Cropland","2021","2021","1000 ha","1129","I","Imputed value","" +"RL","Land Use","728","South Sudan","5110","Area","6620","Cropland","2018","2018","1000 ha","2477.7","I","Imputed value","" +"RL","Land Use","728","South Sudan","5110","Area","6620","Cropland","2019","2019","1000 ha","2477.7","A","Official figure","" +"RL","Land Use","728","South Sudan","5110","Area","6620","Cropland","2020","2020","1000 ha","2477.7","A","Official figure","" +"RL","Land Use","728","South Sudan","5110","Area","6620","Cropland","2021","2021","1000 ha","2479.7","I","Imputed value","" +"RL","Land Use","729","Sudan","5110","Area","6620","Cropland","2018","2018","1000 ha","21210.84","I","Imputed value","" +"RL","Land Use","729","Sudan","5110","Area","6620","Cropland","2019","2019","1000 ha","21210.84","I","Imputed value","" +"RL","Land Use","729","Sudan","5110","Area","6620","Cropland","2020","2020","1000 ha","21210.84","I","Imputed value","" +"RL","Land Use","729","Sudan","5110","Area","6620","Cropland","2021","2021","1000 ha","21210.84","I","Imputed value","" +"RL","Land Use","768","Togo","5110","Area","6620","Cropland","2018","2018","1000 ha","2820","I","Imputed value","" +"RL","Land Use","768","Togo","5110","Area","6620","Cropland","2019","2019","1000 ha","2820","I","Imputed value","" +"RL","Land Use","768","Togo","5110","Area","6620","Cropland","2020","2020","1000 ha","2820","I","Imputed value","" +"RL","Land Use","768","Togo","5110","Area","6620","Cropland","2021","2021","1000 ha","2820","I","Imputed value","" +"RL","Land Use","800","Uganda","5110","Area","6620","Cropland","2018","2018","1000 ha","9100","I","Imputed value","" +"RL","Land Use","800","Uganda","5110","Area","6620","Cropland","2019","2019","1000 ha","9100","I","Imputed value","" +"RL","Land Use","800","Uganda","5110","Area","6620","Cropland","2020","2020","1000 ha","9100","I","Imputed value","" +"RL","Land Use","800","Uganda","5110","Area","6620","Cropland","2021","2021","1000 ha","9100","I","Imputed value","" +"RL","Land Use","834","United Republic of Tanzania","5110","Area","6620","Cropland","2018","2018","1000 ha","15097.4","I","Imputed value","" +"RL","Land Use","834","United Republic of Tanzania","5110","Area","6620","Cropland","2019","2019","1000 ha","15309.7","I","Imputed value","" +"RL","Land Use","834","United Republic of Tanzania","5110","Area","6620","Cropland","2020","2020","1000 ha","15521.2","I","Imputed value","" +"RL","Land Use","834","United Republic of Tanzania","5110","Area","6620","Cropland","2021","2021","1000 ha","15521.2","I","Imputed value","" +"RL","Land Use","894","Zambia","5110","Area","6620","Cropland","2018","2018","1000 ha","3839","I","Imputed value","" +"RL","Land Use","894","Zambia","5110","Area","6620","Cropland","2019","2019","1000 ha","3839","I","Imputed value","" +"RL","Land Use","894","Zambia","5110","Area","6620","Cropland","2020","2020","1000 ha","3839","I","Imputed value","" +"RL","Land Use","894","Zambia","5110","Area","6620","Cropland","2021","2021","1000 ha","3839","I","Imputed value","" +"RL","Land Use","716","Zimbabwe","5110","Area","6620","Cropland","2018","2018","1000 ha","4100","I","Imputed value","" +"RL","Land Use","716","Zimbabwe","5110","Area","6620","Cropland","2019","2019","1000 ha","4100","I","Imputed value","" +"RL","Land Use","716","Zimbabwe","5110","Area","6620","Cropland","2020","2020","1000 ha","4100","I","Imputed value","" +"RL","Land Use","716","Zimbabwe","5110","Area","6620","Cropland","2021","2021","1000 ha","4100","I","Imputed value","" diff --git a/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv new file mode 100644 index 00000000..6da8c5d8 --- /dev/null +++ b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv @@ -0,0 +1,545 @@ +plotid,sampleid,lon,lat,email,flagged,collection_time,analysis_duration,imagery_title,imagery_attributions,sample_geom,pl_sampleid,Does this point correspond to active cropland? +0,0,34.988629829555656,2.403552919494441,aasareansah@gmail.com,false,2023-03-02 16:50,119.3 secs,Planet Monthly Mosaics,,POINT(34.988629829555656 2.403552919494441),0,Non-crop +1,1,40.4369134463712,0.23745156614928092,acgins@umd.edu,false,2023-02-28 19:38,138.7 secs,Planet Monthly Mosaics,,POINT(40.4369134463712 0.237451566149281),1,Non-crop +2,2,35.525785266678824,4.203260690691489,acgins@umd.edu,false,2023-02-28 19:39,54.0 secs,Sentinel-2,,POINT(35.525785266678824 4.203260690691489),2,Non-crop +3,3,35.34502880181526,1.7861325423989396,acgins@umd.edu,false,2023-02-28 19:48,593.3 secs,Sentinel-2,,POINT(35.34502880181526 1.78613254239894),3,Non-crop +4,4,41.092915614480624,-2.088706573827987,acgins@umd.edu,false,2023-03-02 21:56,273.1 secs,Planet Monthly Mosaics,,POINT(41.092915614480624 -2.088706573827987),4,Crop +5,5,34.4249103574835,-1.030559071942206,acgins@umd.edu,false,2023-02-28 20:27,1314.0 secs,Planet Monthly Mosaics,,POINT(34.4249103574835 -1.030559071942206),5,Crop +6,6,39.74092423396898,0.4906842425498013,acgins@umd.edu,false,2023-02-28 20:34,422.6 secs,Planet Monthly Mosaics,,POINT(39.74092423396898 0.490684242549801),6,Non-crop +7,7,35.17727471050547,0.8244599235574998,acgins@umd.edu,false,2023-02-28 20:34,38.1 secs,Sentinel-2,,POINT(35.17727471050547 0.8244599235575),7,Crop +8,8,39.45538305881567,1.1210267423453937,acgins@umd.edu,false,2023-02-28 20:36,88.5 secs,Sentinel-2,,POINT(39.45538305881567 1.121026742345394),8,Non-crop +9,9,39.369209788785,-2.6825226398269617,acgins@umd.edu,false,2023-02-28 20:37,83.3 secs,Planet Monthly Mosaics,,POINT(39.369209788785 -2.682522639826962),9,Non-crop +10,10,34.7476597030811,3.302863534933985,acgins@umd.edu,false,2023-02-28 20:44,382.4 secs,Planet Monthly Mosaics,,POINT(34.7476597030811 3.302863534933985),10,Non-crop +11,11,37.52023980882631,1.1294265556977394,acgins@umd.edu,false,2023-02-28 20:46,111.7 secs,Planet Monthly Mosaics,,POINT(37.52023980882631 1.129426555697739),11,Non-crop +12,12,40.04232638214461,0.10707860224488248,acgins@umd.edu,false,2023-02-28 20:47,82.9 secs,Planet Monthly Mosaics,,POINT(40.04232638214461 0.107078602244882),12,Non-crop +13,13,36.59504709600916,2.079095336922312,acgins@umd.edu,false,2023-02-28 20:48,83.9 secs,Planet Monthly Mosaics,,POINT(36.59504709600916 2.079095336922312),13,Non-crop +14,14,35.355886022749836,-1.7191142947031781,acgins@umd.edu,false,2023-02-28 20:50,83.5 secs,Planet Monthly Mosaics,,POINT(35.355886022749836 -1.719114294703178),14,Non-crop +15,15,37.94355071711686,0.2736254374888986,acgins@umd.edu,false,2023-02-28 20:52,134.4 secs,Planet Monthly Mosaics,,POINT(37.94355071711686 0.273625437488899),15,Crop +16,16,35.567931891106554,-0.963807053974099,acgins@umd.edu,false,2023-02-28 20:54,139.0 secs,Sentinel-2,,POINT(35.567931891106554 -0.963807053974099),16,Crop +17,17,35.2370463385883,0.23539226037375338,acgins@umd.edu,false,2023-02-28 20:55,50.0 secs,Planet Monthly Mosaics,,POINT(35.2370463385883 0.235392260373753),17,Crop +18,18,34.56192216282922,-0.5423313934257865,acgins@umd.edu,false,2023-02-28 21:08,767.9 secs,Mapbox Satellite,,POINT(34.56192216282922 -0.542331393425786),18,Crop +19,19,39.91309533587102,0.5405029647849312,acgins@umd.edu,false,2023-02-28 21:14,378.5 secs,Sentinel-2,,POINT(39.91309533587102 0.540502964784931),19,Non-crop +20,20,37.77622013677315,-3.216077832453689,acgins@umd.edu,false,2023-02-28 21:17,182.3 secs,Mapbox Satellite,,POINT(37.77622013677315 -3.216077832453689),20,Non-crop +21,21,36.952890064644144,-0.7336035276099032,acgins@umd.edu,false,2023-03-02 21:51,1267.9 secs,Planet Monthly Mosaics,,POINT(36.952890064644144 -0.733603527609903),21,Crop +22,22,40.81069538939396,3.8142341143013847,aasareansah@gmail.com,false,2023-03-02 16:52,9.3 secs,Planet Monthly Mosaics,,POINT(40.81069538939396 3.814234114301385),22,Non-crop +23,23,39.31291565564592,-4.292596583003278,aasareansah@gmail.com,false,2023-03-02 16:52,13.0 secs,Planet Monthly Mosaics,,POINT(39.31291565564592 -4.292596583003278),23,Non-crop +24,24,36.56392047775658,-1.4401680504539716,aasareansah@gmail.com,false,2023-03-02 16:53,17.6 secs,Planet Monthly Mosaics,,POINT(36.56392047775658 -1.440168050453972),24,Non-crop +25,25,40.761482926934235,1.797142812318858,aasareansah@gmail.com,false,2023-03-02 16:53,8.4 secs,Planet Monthly Mosaics,,POINT(40.761482926934235 1.797142812318858),25,Non-crop +26,26,35.157061444706144,1.0119532740001498,aasareansah@gmail.com,false,2023-03-02 16:53,44.9 secs,Planet Monthly Mosaics,,POINT(35.157061444706144 1.01195327400015),26,Non-crop +27,27,34.203510614976835,3.843240810122485,aasareansah@gmail.com,false,2023-03-02 16:54,18.9 secs,Planet Monthly Mosaics,,POINT(34.203510614976835 3.843240810122485),27,Non-crop +28,28,40.634491221723714,1.4292896941865394,aasareansah@gmail.com,false,2023-03-02 16:54,11.9 secs,Planet Monthly Mosaics,,POINT(40.634491221723714 1.429289694186539),28,Non-crop +29,29,37.24315858180454,0.18885273302287983,aasareansah@gmail.com,false,2023-03-02 16:54,30.6 secs,Mapbox Satellite,,POINT(37.24315858180454 0.18885273302288),29,Non-crop +30,30,38.50529869815832,1.3266174643205182,aasareansah@gmail.com,false,2023-03-02 16:55,12.2 secs,Planet Monthly Mosaics,,POINT(38.50529869815832 1.326617464320518),30,Non-crop +31,31,40.02570086581538,2.052485246929315,aasareansah@gmail.com,false,2023-03-02 16:55,11.6 secs,Planet Monthly Mosaics,,POINT(40.02570086581538 2.052485246929315),31,Non-crop +32,32,36.08923467501349,2.8152333378024177,aasareansah@gmail.com,false,2023-03-02 16:55,10.4 secs,Planet Monthly Mosaics,,POINT(36.08923467501349 2.815233337802418),32,Non-crop +33,33,38.25468798085764,-3.293961403681881,aasareansah@gmail.com,false,2023-03-02 16:55,7.0 secs,Planet Monthly Mosaics,,POINT(38.25468798085764 -3.293961403681881),33,Non-crop +34,34,36.24856761389988,2.978220395675694,aasareansah@gmail.com,false,2023-03-02 16:55,8.1 secs,Planet Monthly Mosaics,,POINT(36.24856761389988 2.978220395675694),34,Non-crop +35,35,35.08856440679308,-0.610412869850806,aasareansah@gmail.com,false,2023-03-02 16:56,76.0 secs,Planet Monthly Mosaics,,POINT(35.08856440679308 -0.610412869850806),35,Non-crop +36,36,37.10586175136883,0.6450479726392292,aasareansah@gmail.com,false,2023-03-02 16:57,12.4 secs,Planet Monthly Mosaics,,POINT(37.10586175136883 0.645047972639229),36,Non-crop +37,37,36.28157017849977,-0.844978485851314,aasareansah@gmail.com,false,2023-03-02 17:06,575.5 secs,Sentinel-2,,POINT(36.28157017849977 -0.844978485851314),37,Crop +38,38,35.24641350935284,3.1088112936312293,aasareansah@gmail.com,false,2023-03-02 17:06,12.4 secs,Planet Monthly Mosaics,,POINT(35.24641350935284 3.108811293631229),38,Non-crop +39,39,35.725864381252485,-0.9901508750054767,aasareansah@gmail.com,false,2023-03-02 17:07,34.2 secs,Planet Monthly Mosaics,,POINT(35.725864381252485 -0.990150875005477),39,Crop +40,40,36.68065734363421,-0.8149776658973387,aasareansah@gmail.com,false,2023-03-02 17:07,32.3 secs,Planet Monthly Mosaics,,POINT(36.68065734363421 -0.814977665897339),40,Non-crop +41,41,39.78768478044772,3.532698290163748,aasareansah@gmail.com,false,2023-03-02 17:08,13.3 secs,Planet Monthly Mosaics,,POINT(39.78768478044772 3.532698290163748),41,Non-crop +42,42,34.937338523086936,1.1808045765140807,aasareansah@gmail.com,false,2023-03-02 17:08,30.6 secs,Sentinel-2,,POINT(34.937338523086936 1.180804576514081),42,Crop +43,43,35.95117962367605,0.11765502410130428,acgins@umd.edu,false,2023-03-02 22:35,1770.6 secs,Planet Monthly Mosaics,,POINT(35.95117962367605 0.117655024101304),43,Crop +44,44,36.233816518951166,-0.9529835747040568,acgins@umd.edu,false,2023-03-06 22:03,124.3 secs,Mapbox Satellite,,POINT(36.233816518951166 -0.952983574704057),44,Non-crop +45,45,40.34542919363465,-0.16642752812001055,aasareansah@gmail.com,false,2023-03-02 19:02,10.5 secs,Planet Monthly Mosaics,,POINT(40.34542919363465 -0.166427528120011),45,Non-crop +46,46,35.109574883126434,1.7949982414224284,aasareansah@gmail.com,false,2023-03-02 19:02,8.4 secs,Planet Monthly Mosaics,,POINT(35.109574883126434 1.794998241422428),46,Non-crop +47,47,38.10104270259982,2.2770658661008274,aasareansah@gmail.com,false,2023-03-02 19:02,8.7 secs,Planet Monthly Mosaics,,POINT(38.10104270259982 2.277065866100827),47,Non-crop +48,48,34.93916270476336,-0.5632779940672789,aasareansah@gmail.com,false,2023-03-02 19:03,43.4 secs,Planet Monthly Mosaics,,POINT(34.93916270476336 -0.563277994067279),48,Non-crop +49,49,35.698150869472485,2.656619125017031,aasareansah@gmail.com,false,2023-03-02 19:03,7.5 secs,Planet Monthly Mosaics,,POINT(35.698150869472485 2.656619125017031),49,Non-crop +50,50,38.332830459210996,-2.4627477963661004,aasareansah@gmail.com,false,2023-03-02 19:03,7.0 secs,Planet Monthly Mosaics,,POINT(38.332830459210996 -2.4627477963661),50,Non-crop +51,51,36.384083118334274,4.384630880395623,aasareansah@gmail.com,false,2023-03-02 19:03,6.8 secs,Planet Monthly Mosaics,,POINT(36.384083118334274 4.384630880395623),51,Non-crop +52,52,35.01471998180397,-0.42170819623595546,acgins@umd.edu,false,2023-03-02 22:38,26.5 secs,Planet Monthly Mosaics,,POINT(35.01471998180397 -0.421708196235955),52,Crop +53,53,37.47142387421962,2.60955944432736,aasareansah@gmail.com,false,2023-03-02 19:04,7.2 secs,Planet Monthly Mosaics,,POINT(37.47142387421962 2.60955944432736),53,Non-crop +54,54,37.44533838762773,-1.8552448087267317,aasareansah@gmail.com,false,2023-03-02 19:04,16.4 secs,Mapbox Satellite,,POINT(37.44533838762773 -1.855244808726732),54,Non-crop +55,55,40.14514392119623,-3.098656053349216,aasareansah@gmail.com,false,2023-03-02 19:04,7.7 secs,Mapbox Satellite,,POINT(40.14514392119623 -3.098656053349216),55,Non-crop +56,56,38.6594746226374,-3.835613952172189,aasareansah@gmail.com,false,2023-03-02 19:04,18.0 secs,Planet Monthly Mosaics,,POINT(38.6594746226374 -3.835613952172189),56,Non-crop +57,57,39.02562274024034,0.07077064319861262,aasareansah@gmail.com,false,2023-03-02 19:05,7.7 secs,Planet Monthly Mosaics,,POINT(39.02562274024034 0.070770643198613),57,Non-crop +58,58,36.02109758574353,3.5366108060174963,aasareansah@gmail.com,false,2023-03-02 19:05,9.3 secs,Planet Monthly Mosaics,,POINT(36.02109758574353 3.536610806017496),58,Non-crop +59,59,35.06865100448405,1.187649341007699,aasareansah@gmail.com,false,2023-03-02 19:05,18.5 secs,Planet Monthly Mosaics,,POINT(35.06865100448405 1.187649341007699),59,Crop +60,60,38.263410192690564,2.051383658830262,aasareansah@gmail.com,false,2023-03-02 19:05,8.3 secs,Planet Monthly Mosaics,,POINT(38.263410192690564 2.051383658830262),60,Non-crop +61,61,40.14898007195484,3.3226177399453816,aasareansah@gmail.com,false,2023-03-02 19:05,17.1 secs,Planet Monthly Mosaics,,POINT(40.14898007195484 3.322617739945382),61,Non-crop +62,62,37.206652528458754,-1.1530997062886132,acgins@umd.edu,false,2023-03-02 22:41,192.7 secs,Planet Monthly Mosaics,,POINT(37.206652528458754 -1.153099706288613),62,Crop +63,63,36.33551213412802,-2.0059902859724703,acgins@umd.edu,false,2023-03-02 22:42,83.8 secs,Planet Monthly Mosaics,,POINT(36.33551213412802 -2.00599028597247),63,Non-crop +64,64,37.81898292849584,2.890470600118884,acgins@umd.edu,false,2023-03-02 22:43,32.4 secs,Mapbox Satellite,,POINT(37.81898292849584 2.890470600118884),64,Non-crop +65,65,36.964503175730776,-1.8728601229685733,acgins@umd.edu,false,2023-03-06 22:06,178.2 secs,Mapbox Satellite,,POINT(36.964503175730776 -1.872860122968573),65,Non-crop +66,66,35.432462250513026,-0.21706777965244287,acgins@umd.edu,false,2023-03-02 22:45,49.9 secs,Sentinel-2,,POINT(35.432462250513026 -0.217067779652443),66,Crop +67,67,35.12381890958649,0.7606469113885553,acgins@umd.edu,false,2023-03-06 22:09,180.7 secs,Mapbox Satellite,,POINT(35.12381890958649 0.760646911388555),67,Crop +68,68,34.30929368936593,0.47748228001945475,acgins@umd.edu,false,2023-03-02 22:54,215.1 secs,Mapbox Satellite,,POINT(34.30929368936593 0.477482280019455),68,Crop +69,69,36.6457343594037,-0.49640885252982253,acgins@umd.edu,false,2023-03-02 22:54,35.7 secs,Planet Monthly Mosaics,,POINT(36.6457343594037 -0.496408852529823),69,Non-crop +70,70,34.226388257459824,-0.6041420114577926,acgins@umd.edu,false,2023-03-02 22:56,107.2 secs,Planet Monthly Mosaics,,POINT(34.226388257459824 -0.604142011457793),70,Crop +71,71,38.249558852614086,3.1963730568049713,acgins@umd.edu,false,2023-03-02 22:57,38.0 secs,Planet Monthly Mosaics,,POINT(38.249558852614086 3.196373056804971),71,Non-crop +72,72,35.416853560011795,3.8915106166058218,acgins@umd.edu,false,2023-03-02 22:57,21.4 secs,Planet Monthly Mosaics,,POINT(35.416853560011795 3.891510616605822),72,Non-crop +73,73,34.793844712321096,1.0897067468642034,acgins@umd.edu,false,2023-03-02 22:59,92.3 secs,Planet Monthly Mosaics,,POINT(34.793844712321096 1.089706746864203),73,Crop +74,74,35.82894420505845,0.4963590966445787,acgins@umd.edu,false,2023-03-06 22:09,7.4 secs,Mapbox Satellite,,POINT(35.82894420505845 0.496359096644579),74,Non-crop +75,75,40.86680566262454,2.5510833756481532,acgins@umd.edu,false,2023-03-02 23:01,67.9 secs,Planet Monthly Mosaics,,POINT(40.86680566262454 2.551083375648153),75,Non-crop +76,76,37.42434808348648,2.172412965946544,acgins@umd.edu,false,2023-03-02 23:01,10.0 secs,Planet Monthly Mosaics,,POINT(37.42434808348648 2.172412965946544),76,Non-crop +77,77,37.15129430727518,-0.7226093010817065,acgins@umd.edu,false,2023-03-06 22:11,95.6 secs,Mapbox Satellite,,POINT(37.15129430727518 -0.722609301081706),77,Non-crop +78,78,35.24400762282487,-0.8344837781524341,acgins@umd.edu,false,2023-03-02 23:03,54.7 secs,Sentinel-2,,POINT(35.24400762282487 -0.834483778152434),78,Crop +79,79,35.89383258961963,-0.3086155983480622,acgins@umd.edu,false,2023-03-04 19:09,25.1 secs,Planet Monthly Mosaics,,POINT(35.89383258961963 -0.308615598348062),79,Crop +80,80,34.8374317772852,1.2105020095980699,acgins@umd.edu,false,2023-03-04 19:10,16.7 secs,Planet Monthly Mosaics,,POINT(34.8374317772852 1.21050200959807),80,Crop +81,81,37.47851943218349,-1.5673861189284093,acgins@umd.edu,false,2023-03-04 19:13,182.4 secs,Planet Monthly Mosaics,,POINT(37.47851943218349 -1.567386118928409),81,Crop +82,82,37.62528047925142,-3.386462077890213,acgins@umd.edu,false,2023-03-04 19:14,58.1 secs,Planet Monthly Mosaics,,POINT(37.62528047925142 -3.386462077890213),82,Crop +83,83,37.94051901411475,0.003977377954060247,acgins@umd.edu,false,2023-03-04 19:14,43.1 secs,Mapbox Satellite,,POINT(37.94051901411475 0.00397737795406),83,Crop +84,84,38.7070591354366,-2.0059845289306772,acgins@umd.edu,false,2023-03-04 19:15,22.4 secs,Planet Monthly Mosaics,,POINT(38.7070591354366 -2.005984528930677),84,Non-crop +85,85,34.838779927856656,-0.86257077504155,acgins@umd.edu,false,2023-03-04 19:19,285.8 secs,Planet Monthly Mosaics,,POINT(34.838779927856656 -0.86257077504155),85,Crop +86,86,37.44660412263658,1.2193124117332885,acgins@umd.edu,false,2023-03-04 19:20,16.2 secs,Planet Monthly Mosaics,,POINT(37.44660412263658 1.219312411733288),86,Non-crop +87,87,34.6632601985172,0.07559985882881662,acgins@umd.edu,false,2023-03-04 19:22,117.0 secs,Planet Monthly Mosaics,,POINT(34.6632601985172 0.075599858828817),87,Non-crop +88,88,34.638949502003825,0.6729112014973999,acgins@umd.edu,false,2023-03-04 19:22,20.9 secs,Planet Monthly Mosaics,,POINT(34.638949502003825 0.6729112014974),88,Crop +89,89,34.5994023707549,0.7630158568229832,acgins@umd.edu,false,2023-03-04 19:23,35.1 secs,Planet Monthly Mosaics,,POINT(34.5994023707549 0.763015856822983),89,Crop +90,90,36.93706274251377,2.255505688906279,acgins@umd.edu,false,2023-03-04 19:23,39.5 secs,Mapbox Satellite,,POINT(36.93706274251377 2.255505688906279),90,Non-crop +91,91,40.44223583641046,4.03082660667592,acgins@umd.edu,false,2023-03-04 19:24,26.3 secs,Planet Monthly Mosaics,,POINT(40.44223583641046 4.03082660667592),91,Non-crop +92,92,36.06039300271393,-0.5684581218619031,acgins@umd.edu,false,2023-03-04 19:24,32.5 secs,Mapbox Satellite,,POINT(36.06039300271393 -0.568458121861903),92,Crop +93,93,37.59588686792248,0.11931919114708699,acgins@umd.edu,false,2023-03-04 19:25,33.1 secs,Planet Monthly Mosaics,,POINT(37.59588686792248 0.119319191147087),93,Non-crop +94,94,34.95006733497294,-0.7711616777128759,acgins@umd.edu,false,2023-03-06 22:16,311.0 secs,Planet Monthly Mosaics,,POINT(34.95006733497294 -0.771161677712876),94,Crop +95,95,34.37280204804135,-0.5213450533332684,acgins@umd.edu,false,2023-03-06 21:13,139.4 secs,,,POINT(34.37280204804135 -0.521345053333268),95,Crop +96,96,35.92793233525755,-2.0132818618461816,acgins@umd.edu,false,2023-03-06 21:13,35.3 secs,Planet Monthly Mosaics,,POINT(35.92793233525755 -2.013281861846182),96,Non-crop +97,97,40.50489472621563,-1.7350862750790679,acgins@umd.edu,false,2023-03-06 21:14,53.6 secs,Planet Monthly Mosaics,,POINT(40.50489472621563 -1.735086275079068),97,Non-crop +98,98,34.84343325317191,-0.5513431781255153,acgins@umd.edu,false,2023-03-06 21:15,41.8 secs,Planet Monthly Mosaics,,POINT(34.84343325317191 -0.551343178125515),98,Crop +99,99,38.810924566797794,2.4564363287855104,acgins@umd.edu,false,2023-03-06 21:15,24.7 secs,Planet Monthly Mosaics,,POINT(38.810924566797794 2.45643632878551),99,Non-crop +100,100,38.82801401094352,0.44583460966597566,acgins@umd.edu,false,2023-03-06 21:16,15.8 secs,Planet Monthly Mosaics,,POINT(38.82801401094352 0.445834609665976),100,Non-crop +101,101,35.01842360551604,1.951978852908563,acgins@umd.edu,false,2023-03-06 21:16,19.8 secs,Planet Monthly Mosaics,,POINT(35.01842360551604 1.951978852908563),101,Non-crop +102,102,38.63846258841169,-1.8744628832696686,acgins@umd.edu,false,2023-03-06 21:17,82.1 secs,Planet Monthly Mosaics,,POINT(38.63846258841169 -1.874462883269669),102,Non-crop +103,103,34.67637175915808,4.509081535107014,acgins@umd.edu,false,2023-03-06 21:18,35.9 secs,Planet Monthly Mosaics,,POINT(34.67637175915808 4.509081535107014),103,Non-crop +104,104,34.80059714936316,-0.6330272098164963,acgins@umd.edu,false,2023-03-06 21:20,94.9 secs,Planet Monthly Mosaics,,POINT(34.80059714936316 -0.633027209816496),104,Crop +105,105,39.03989200445923,-2.346187888919231,acgins@umd.edu,false,2023-03-06 21:21,83.9 secs,Planet Monthly Mosaics,,POINT(39.03989200445923 -2.346187888919231),105,Non-crop +106,106,37.75101158485352,-0.05595061740159739,acgins@umd.edu,false,2023-03-06 21:23,131.6 secs,Planet Monthly Mosaics,,POINT(37.75101158485352 -0.055950617401597),106,Crop +107,107,34.94705324098189,-0.5166194835899087,acgins@umd.edu,false,2023-03-06 21:24,49.1 secs,Planet Monthly Mosaics,,POINT(34.94705324098189 -0.516619483589909),107,Crop +108,108,36.43853260211107,3.97061768786652,acgins@umd.edu,false,2023-03-06 21:24,9.9 secs,Mapbox Satellite,,POINT(36.43853260211107 3.97061768786652),108,Non-crop +109,109,38.8861466691318,3.0445834421692615,acgins@umd.edu,false,2023-03-06 21:25,23.7 secs,Planet Monthly Mosaics,,POINT(38.8861466691318 3.044583442169262),109,Non-crop +110,110,35.876614895369954,3.7678862374880135,acgins@umd.edu,false,2023-03-06 21:25,33.9 secs,Planet Monthly Mosaics,,POINT(35.876614895369954 3.767886237488014),110,Non-crop +111,111,38.82883825639645,2.5306209549193475,acgins@umd.edu,false,2023-03-06 21:33,496.2 secs,Planet Monthly Mosaics,,POINT(38.82883825639645 2.530620954919348),111,Non-crop +112,112,38.2367270952654,-1.3301300190469625,acgins@umd.edu,false,2023-03-06 22:17,55.9 secs,Mapbox Satellite,,POINT(38.2367270952654 -1.330130019046962),112,Non-crop +113,113,37.63554202932443,0.03299322115736923,acgins@umd.edu,false,2023-03-06 21:39,216.8 secs,Planet Monthly Mosaics,,POINT(37.63554202932443 0.032993221157369),113,Crop +114,114,37.05135107858711,-0.388438674158326,acgins@umd.edu,false,2023-03-06 21:41,138.7 secs,Planet Monthly Mosaics,,POINT(37.05135107858711 -0.388438674158326),114,Crop +115,115,38.39683562140306,-3.562375659885144,acgins@umd.edu,false,2023-03-06 21:42,34.6 secs,Planet Monthly Mosaics,,POINT(38.39683562140306 -3.562375659885144),115,Crop +116,116,38.47807537401632,-1.3194672935579193,acgins@umd.edu,false,2023-03-06 22:23,189.1 secs,Planet Monthly Mosaics,,POINT(38.47807537401632 -1.319467293557919),116,Non-crop +117,117,37.90724334671853,1.2126009407086038,acgins@umd.edu,false,2023-03-06 21:45,8.3 secs,Planet Monthly Mosaics,,POINT(37.90724334671853 1.212600940708604),117,Non-crop +118,118,35.1493469381087,0.7857613956282058,acgins@umd.edu,false,2023-03-06 21:47,87.4 secs,Planet Monthly Mosaics,,POINT(35.1493469381087 0.785761395628206),118,Crop +119,119,37.04107541888969,2.415639616416417,acgins@umd.edu,false,2023-03-06 21:47,40.4 secs,Planet Monthly Mosaics,,POINT(37.04107541888969 2.415639616416417),119,Non-crop +120,120,37.135914893319935,0.8892938871269861,acgins@umd.edu,false,2023-03-06 21:48,8.6 secs,Planet Monthly Mosaics,,POINT(37.135914893319935 0.889293887126986),120,Non-crop +121,121,34.280251153267145,-0.8857188443090579,acgins@umd.edu,false,2023-03-06 21:49,90.0 secs,Sentinel-2,,POINT(34.280251153267145 -0.885718844309058),121,Crop +122,122,37.104919656283776,0.944772577359767,acgins@umd.edu,false,2023-03-06 21:50,37.8 secs,Planet Monthly Mosaics,,POINT(37.104919656283776 0.944772577359767),122,Non-crop +123,123,36.91642788302333,-1.8734338149965883,acgins@umd.edu,false,2023-03-06 21:50,28.5 secs,Mapbox Satellite,,POINT(36.91642788302333 -1.873433814996588),123,Non-crop +124,124,35.6530472090822,1.3337481691783033,acgins@umd.edu,false,2023-03-06 21:53,188.2 secs,Planet Monthly Mosaics,,POINT(35.6530472090822 1.333748169178303),124,Crop +125,125,37.665284200034286,-0.8743618235367431,acgins@umd.edu,false,2023-03-06 21:56,174.0 secs,Sentinel-2,,POINT(37.665284200034286 -0.874361823536743),125,Non-crop +126,126,40.28931527278772,-1.6826194889901331,acgins@umd.edu,false,2023-03-06 21:57,33.0 secs,Planet Monthly Mosaics,,POINT(40.28931527278772 -1.682619488990133),126,Non-crop +127,127,35.24886337281288,0.9780628088660805,acgins@umd.edu,false,2023-03-06 21:57,15.6 secs,Planet Monthly Mosaics,,POINT(35.24886337281288 0.97806280886608),127,Crop +128,128,35.350804743305915,-0.14292567584629862,acgins@umd.edu,false,2023-03-06 21:59,99.4 secs,Planet Monthly Mosaics,,POINT(35.350804743305915 -0.142925675846299),128,Crop +129,129,36.875372636897794,-1.9178813516888036,acgins@umd.edu,false,2023-03-06 21:59,37.4 secs,Planet Monthly Mosaics,,POINT(36.875372636897794 -1.917881351688804),129,Non-crop +130,130,36.36696602805077,-1.0011713498446717,acgins@umd.edu,false,2023-03-06 22:01,87.2 secs,Mapbox Satellite,,POINT(36.36696602805077 -1.001171349844672),130,Non-crop +131,131,34.03567997149679,-0.45059458138419733,acgins@umd.edu,false,2023-03-06 22:30,453.4 secs,Planet Monthly Mosaics,,POINT(34.03567997149679 -0.450594581384197),131,Non-crop +132,132,34.31052408060243,0.06011131997460628,acgins@umd.edu,false,2023-03-06 22:31,21.8 secs,Mapbox Satellite,,POINT(34.31052408060243 0.060111319974606),132,Non-crop +133,133,40.112200611852735,3.057621068313114,acgins@umd.edu,false,2023-03-06 22:31,20.7 secs,Mapbox Satellite,,POINT(40.112200611852735 3.057621068313114),133,Non-crop +134,134,37.59446156272681,-0.6348207832764996,acgins@umd.edu,false,2023-03-06 22:38,440.2 secs,Planet Monthly Mosaics,,POINT(37.59446156272681 -0.6348207832765),134,Non-crop +135,135,40.9961063143978,-1.6779669321206159,acgins@umd.edu,false,2023-03-06 22:46,481.6 secs,Planet Monthly Mosaics,,POINT(40.9961063143978 -1.677966932120616),135,Non-crop +136,136,39.22212647015637,2.6922268323855563,acgins@umd.edu,false,2023-03-06 22:47,40.9 secs,Planet Monthly Mosaics,,POINT(39.22212647015637 2.692226832385556),136,Non-crop +137,137,36.21832154143681,-0.33166413329927724,acgins@umd.edu,false,2023-03-06 22:48,43.8 secs,Planet Monthly Mosaics,,POINT(36.21832154143681 -0.331664133299277),137,Crop +138,138,34.51475222889342,-0.7822238210290815,acgins@umd.edu,false,2023-03-06 22:48,35.5 secs,Planet Monthly Mosaics,,POINT(34.51475222889342 -0.782223821029082),138,Crop +139,139,38.50386682037691,-0.7440523952127349,acgins@umd.edu,false,2023-03-06 22:49,29.5 secs,Planet Monthly Mosaics,,POINT(38.50386682037691 -0.744052395212735),139,Non-crop +140,140,34.544980328359806,3.3127536927240926,acgins@umd.edu,false,2023-03-06 22:49,23.1 secs,Mapbox Satellite,,POINT(34.544980328359806 3.312753692724093),140,Non-crop +141,141,35.072614057925755,2.0515640292935595,acgins@umd.edu,false,2023-03-06 22:50,31.9 secs,Planet Monthly Mosaics,,POINT(35.072614057925755 2.05156402929356),141,Non-crop +142,142,37.63731851488094,-0.7082741543121748,acgins@umd.edu,false,2023-03-06 22:52,110.3 secs,Mapbox Satellite,,POINT(37.63731851488094 -0.708274154312175),142,Crop +143,143,36.79356032174582,-1.3320484579772995,acgins@umd.edu,false,2023-03-06 22:52,12.9 secs,Mapbox Satellite,,POINT(36.79356032174582 -1.3320484579773),143,Non-crop +144,144,37.196987405380554,-0.997502551481028,acgins@umd.edu,false,2023-03-06 22:52,29.5 secs,Planet Monthly Mosaics,,POINT(37.196987405380554 -0.997502551481028),144,Crop +145,145,35.97558762159962,-0.12276447907124412,acgins@umd.edu,false,2023-03-06 22:53,16.7 secs,Planet Monthly Mosaics,,POINT(35.97558762159962 -0.122764479071244),145,Crop +146,146,34.1891721541313,-0.6146176829171154,acgins@umd.edu,false,2023-03-06 22:54,105.1 secs,Planet Monthly Mosaics,,POINT(34.1891721541313 -0.614617682917115),146,Crop +147,147,36.92847527942525,1.2782331263275961,acgins@umd.edu,false,2023-03-06 22:55,21.4 secs,Planet Monthly Mosaics,,POINT(36.92847527942525 1.278233126327596),147,Non-crop +148,148,37.67795253138457,-0.29646401880195755,acgins@umd.edu,false,2023-03-06 22:56,100.1 secs,Mapbox Satellite,,POINT(37.67795253138457 -0.296464018801958),148,Crop +149,149,38.06462053176975,-1.043559942579206,acgins@umd.edu,false,2023-03-06 22:57,48.4 secs,Planet Monthly Mosaics,,POINT(38.06462053176975 -1.043559942579206),149,Crop +150,150,34.886305931598756,1.5162521774427242,acgins@umd.edu,false,2023-03-06 23:08,628.7 secs,Planet Monthly Mosaics,,POINT(34.886305931598756 1.516252177442724),150,Non-crop +151,151,36.96408686642167,0.4599662921332445,acgins@umd.edu,false,2023-03-06 23:09,51.0 secs,Planet Monthly Mosaics,,POINT(36.96408686642167 0.459966292133244),151,Non-crop +152,152,35.13520286056352,-1.1297140056307589,acgins@umd.edu,false,2023-03-06 23:10,92.2 secs,Planet Monthly Mosaics,,POINT(35.13520286056352 -1.129714005630759),152,Crop +153,153,41.800681225973825,3.9420204760538824,acgins@umd.edu,false,2023-03-06 23:11,32.5 secs,Planet Monthly Mosaics,,POINT(41.800681225973825 3.942020476053882),153,Non-crop +154,154,36.41868031095638,-1.6763774091068298,acgins@umd.edu,false,2023-03-06 23:11,38.6 secs,Planet Monthly Mosaics,,POINT(36.41868031095638 -1.67637740910683),154,Non-crop +155,155,38.033900036116925,-1.3282559996083236,acgins@umd.edu,false,2023-03-06 23:16,278.6 secs,Planet Monthly Mosaics,,POINT(38.033900036116925 -1.328255999608324),155,Non-crop +156,156,39.29015146450414,-4.3996825162698165,acgins@umd.edu,false,2023-03-06 23:19,187.0 secs,Sentinel-2,,POINT(39.29015146450414 -4.399682516269816),156,Crop +157,157,37.44623545167376,-1.0728680549928664,acgins@umd.edu,false,2023-03-06 23:21,100.7 secs,Planet Monthly Mosaics,,POINT(37.44623545167376 -1.072868054992866),157,Non-crop +158,158,40.00198076172898,-1.4034050963748328,acgins@umd.edu,false,2023-03-06 23:21,11.9 secs,Planet Monthly Mosaics,,POINT(40.00198076172898 -1.403405096374833),158,Crop +159,159,34.51540863900097,0.20085040451937342,acgins@umd.edu,false,2023-03-06 23:21,8.2 secs,Planet Monthly Mosaics,,POINT(34.51540863900097 0.200850404519373),159,Non-crop +160,160,34.364068159964674,-0.027030376910205087,acgins@umd.edu,false,2023-03-06 23:22,27.7 secs,Planet Monthly Mosaics,,POINT(34.364068159964674 -0.027030376910205),160,Crop +161,161,38.97895604283259,2.4386926635852486,acgins@umd.edu,false,2023-03-06 23:22,10.7 secs,Planet Monthly Mosaics,,POINT(38.97895604283259 2.438692663585249),161,Non-crop +162,162,35.27387999609261,0.11199548140903959,acgins@umd.edu,false,2023-03-06 23:22,19.9 secs,Planet Monthly Mosaics,,POINT(35.27387999609261 0.11199548140904),162,Crop +163,163,39.911542341177416,-2.5611092030692055,acgins@umd.edu,false,2023-03-06 23:22,27.7 secs,Mapbox Satellite,,POINT(39.911542341177416 -2.561109203069206),163,Non-crop +164,164,35.185079473396854,0.5617606376305225,acgins@umd.edu,false,2023-03-06 23:23,16.7 secs,Planet Monthly Mosaics,,POINT(35.185079473396854 0.561760637630522),164,Crop +165,165,35.05688885441114,-0.6795407745893438,acgins@umd.edu,false,2023-03-06 23:25,152.2 secs,Planet Monthly Mosaics,,POINT(35.05688885441114 -0.679540774589344),165,Crop +166,166,37.58793262053442,3.7690288499729587,acgins@umd.edu,false,2023-03-06 23:26,14.9 secs,Mapbox Satellite,,POINT(37.58793262053442 3.769028849972959),166,Non-crop +167,167,34.45875563125551,0.6272795031255634,acgins@umd.edu,false,2023-03-06 23:26,40.6 secs,Mapbox Satellite,,POINT(34.45875563125551 0.627279503125563),167,Crop +168,168,34.40926954593081,0.2551783003635253,aasareansah@gmail.com,false,2023-03-07 15:02,36.9 secs,Planet Monthly Mosaics,,POINT(34.40926954593081 0.255178300363525),168,Non-crop +169,169,39.660802154481644,-3.190642695064495,aasareansah@gmail.com,false,2023-03-07 15:02,13.8 secs,Planet Monthly Mosaics,,POINT(39.660802154481644 -3.190642695064495),169,Non-crop +170,170,34.54046798139366,0.6859414994541169,aasareansah@gmail.com,false,2023-03-07 15:03,32.4 secs,Planet Monthly Mosaics,,POINT(34.54046798139366 0.685941499454117),170,Crop +171,171,39.073614246897385,-0.2633723289888229,aasareansah@gmail.com,false,2023-03-07 15:03,12.4 secs,Planet Monthly Mosaics,,POINT(39.073614246897385 -0.263372328988823),171,Non-crop +172,172,38.432919130857414,1.7396830042557871,aasareansah@gmail.com,false,2023-03-07 15:03,9.2 secs,Planet Monthly Mosaics,,POINT(38.432919130857414 1.739683004255787),172,Non-crop +173,173,37.095575381750336,-1.0096069277601003,aasareansah@gmail.com,false,2023-03-07 15:03,15.0 secs,Planet Monthly Mosaics,,POINT(37.095575381750336 -1.0096069277601),173,Non-crop +174,174,36.65947410135244,-1.1671998161925197,aasareansah@gmail.com,false,2023-03-07 15:04,24.3 secs,Planet Monthly Mosaics,,POINT(36.65947410135244 -1.16719981619252),174,Crop +175,175,34.45338777219258,-0.7355236000562643,aasareansah@gmail.com,false,2023-03-07 15:04,9.5 secs,Planet Monthly Mosaics,,POINT(34.45338777219258 -0.735523600056264),175,Non-crop +176,176,35.91774080987921,-1.0793531043719524,aasareansah@gmail.com,false,2023-03-07 15:04,13.8 secs,Planet Monthly Mosaics,,POINT(35.91774080987921 -1.079353104371952),176,Non-crop +177,177,34.72286953140142,0.4356879558642015,aasareansah@gmail.com,false,2023-03-07 15:04,14.1 secs,Planet Monthly Mosaics,,POINT(34.72286953140142 0.435687955864202),177,Non-crop +178,178,35.247612435415604,1.055323072725927,aasareansah@gmail.com,false,2023-03-07 15:04,19.3 secs,Planet Monthly Mosaics,,POINT(35.247612435415604 1.055323072725927),178,Crop +179,179,34.17397768556119,0.41298920794555216,aasareansah@gmail.com,false,2023-03-07 15:05,14.7 secs,Planet Monthly Mosaics,,POINT(34.17397768556119 0.412989207945552),179,Non-crop +180,180,36.90739764160443,-0.6045908967298282,aasareansah@gmail.com,false,2023-03-07 15:05,18.5 secs,Planet Monthly Mosaics,,POINT(36.90739764160443 -0.604590896729828),180,Non-crop +181,181,35.09232298964106,-0.24405121259209608,aasareansah@gmail.com,false,2023-03-07 15:06,66.5 secs,Planet Monthly Mosaics,,POINT(35.09232298964106 -0.244051212592096),181,Non-crop +182,182,34.93320904860172,3.089392923071744,aasareansah@gmail.com,false,2023-03-07 15:06,15.0 secs,Planet Monthly Mosaics,,POINT(34.93320904860172 3.089392923071744),182,Non-crop +183,183,35.386016558505396,0.7061819098528996,aasareansah@gmail.com,false,2023-03-07 15:06,20.2 secs,Planet Monthly Mosaics,,POINT(35.386016558505396 0.7061819098529),183,Crop +184,184,35.44600492019322,0.6737515821752349,acgins@umd.edu,false,2023-03-07 20:48,266.8 secs,Planet Monthly Mosaics,,POINT(35.44600492019322 0.673751582175235),184,Non-crop +185,185,36.83715746471805,-0.3040771550035826,aasareansah@gmail.com,false,2023-03-07 15:08,22.3 secs,Planet Monthly Mosaics,,POINT(36.83715746471805 -0.304077155003583),185,Crop +186,186,35.47235817550058,-1.7747537021443485,aasareansah@gmail.com,false,2023-03-07 15:08,9.6 secs,Planet Monthly Mosaics,,POINT(35.47235817550058 -1.774753702144348),186,Non-crop +187,187,35.9969624367669,-0.05057581718672911,acgins@umd.edu,false,2023-03-07 20:57,561.7 secs,Planet Monthly Mosaics,,POINT(35.9969624367669 -0.050575817186729),187,Crop +188,188,35.02138764947741,0.9068522061088463,aasareansah@gmail.com,false,2023-03-07 15:09,16.1 secs,Planet Monthly Mosaics,,POINT(35.02138764947741 0.906852206108846),188,Non-crop +189,189,37.28290916070205,-0.5670627725341534,acgins@umd.edu,false,2023-03-07 21:03,314.8 secs,Planet Monthly Mosaics,,POINT(37.28290916070205 -0.567062772534153),189,Non-crop +190,190,36.27531089330441,1.6312781626526536,aasareansah@gmail.com,false,2023-03-07 15:11,9.3 secs,Planet Monthly Mosaics,,POINT(36.27531089330441 1.631278162652654),190,Non-crop +191,191,38.18613982888146,2.3095627412092896,aasareansah@gmail.com,false,2023-03-07 15:11,11.6 secs,Planet Monthly Mosaics,,POINT(38.18613982888146 2.30956274120929),191,Non-crop +192,192,38.82823907086845,2.961020075278439,aasareansah@gmail.com,false,2023-03-07 15:11,10.1 secs,Planet Monthly Mosaics,,POINT(38.82823907086845 2.961020075278439),192,Non-crop +193,193,35.756354105431015,-0.25214860600974354,aasareansah@gmail.com,false,2023-03-07 15:11,14.7 secs,Planet Monthly Mosaics,,POINT(35.756354105431015 -0.252148606009744),193,Crop +194,194,34.05911892769959,-0.7668905312894166,aasareansah@gmail.com,false,2023-03-07 15:12,38.0 secs,Planet Monthly Mosaics,,POINT(34.05911892769959 -0.766890531289417),194,Non-crop +195,195,37.434398228197466,-1.4124646108398355,aasareansah@gmail.com,false,2023-03-07 15:12,16.4 secs,Planet Monthly Mosaics,,POINT(37.434398228197466 -1.412464610839836),195,Non-crop +196,196,34.24813006085403,-0.18754586249713845,aasareansah@gmail.com,false,2023-03-07 15:13,19.2 secs,Planet Monthly Mosaics,,POINT(34.24813006085403 -0.187545862497138),196,Non-crop +197,197,39.38163103579628,-0.5870464083569815,aasareansah@gmail.com,false,2023-03-07 15:13,13.3 secs,Planet Monthly Mosaics,,POINT(39.38163103579628 -0.587046408356982),197,Non-crop +198,198,34.407612059311866,4.020736333189546,aasareansah@gmail.com,false,2023-03-07 15:13,9.6 secs,Planet Monthly Mosaics,,POINT(34.407612059311866 4.020736333189546),198,Non-crop +199,199,34.5106962080211,-0.1159075048798257,acgins@umd.edu,false,2023-03-07 21:04,62.6 secs,Mapbox Satellite,,POINT(34.5106962080211 -0.115907504879826),199,Crop +200,200,39.699070498363014,0.8268795354015316,aasareansah@gmail.com,false,2023-03-07 15:14,17.0 secs,Planet Monthly Mosaics,,POINT(39.699070498363014 0.826879535401532),200,Non-crop +201,201,35.26606385102699,-0.8488962562008477,acgins@umd.edu,false,2023-03-07 21:09,306.0 secs,Planet Monthly Mosaics,,POINT(35.26606385102699 -0.848896256200848),201,Crop +202,202,37.00569584470588,-0.4607649473863657,aasareansah@gmail.com,false,2023-03-07 15:15,26.7 secs,Planet Monthly Mosaics,,POINT(37.00569584470588 -0.460764947386366),202,Non-crop +203,203,34.681833787046784,-0.0017856800434060153,aasareansah@gmail.com,false,2023-03-07 15:16,56.4 secs,Planet Monthly Mosaics,,POINT(34.681833787046784 -0.001785680043406),203,Non-crop +204,204,35.00322143113498,-0.37269709187609834,aasareansah@gmail.com,false,2023-03-07 15:16,14.3 secs,Planet Monthly Mosaics,,POINT(35.00322143113498 -0.372697091876098),204,Non-crop +205,205,35.09439243911183,0.4304058574197333,aasareansah@gmail.com,false,2023-03-07 15:17,62.4 secs,Planet Monthly Mosaics,,POINT(35.09439243911183 0.430405857419733),205,Crop +206,206,35.13586471270375,3.8666126635096743,aasareansah@gmail.com,false,2023-03-07 15:17,13.6 secs,Planet Monthly Mosaics,,POINT(35.13586471270375 3.866612663509674),206,Non-crop +207,207,38.15212804664146,3.3677157950076575,aasareansah@gmail.com,false,2023-03-07 15:17,9.7 secs,Planet Monthly Mosaics,,POINT(38.15212804664146 3.367715795007658),207,Non-crop +208,208,38.286578480177624,-3.580862928287623,aasareansah@gmail.com,false,2023-03-07 15:17,8.5 secs,Planet Monthly Mosaics,,POINT(38.286578480177624 -3.580862928287623),208,Non-crop +209,209,39.404093137783995,2.45152108353291,aasareansah@gmail.com,false,2023-03-07 15:17,11.3 secs,Planet Monthly Mosaics,,POINT(39.404093137783995 2.45152108353291),209,Non-crop +210,210,35.065711700959476,2.547656684794862,aasareansah@gmail.com,false,2023-03-07 15:18,9.6 secs,Planet Monthly Mosaics,,POINT(35.065711700959476 2.547656684794862),210,Non-crop +211,211,35.947769272904495,0.20415397808837918,aasareansah@gmail.com,false,2023-03-07 15:18,21.9 secs,Planet Monthly Mosaics,,POINT(35.947769272904495 0.204153978088379),211,Non-crop +212,212,37.91801786296307,-2.1375884435327785,aasareansah@gmail.com,false,2023-03-07 15:19,51.9 secs,Planet Monthly Mosaics,,POINT(37.91801786296307 -2.137588443532778),212,Non-crop +213,213,36.88669703425683,0.42706564857485335,aasareansah@gmail.com,false,2023-03-07 15:19,9.7 secs,Planet Monthly Mosaics,,POINT(36.88669703425683 0.427065648574853),213,Non-crop +214,214,34.068160039789994,-0.6537499847545581,aasareansah@gmail.com,false,2023-03-07 15:19,15.2 secs,Planet Monthly Mosaics,,POINT(34.068160039789994 -0.653749984754558),214,Crop +215,215,37.99915424752388,0.06625355062429386,aasareansah@gmail.com,false,2023-03-07 15:20,34.6 secs,Planet Monthly Mosaics,,POINT(37.99915424752388 0.066253550624294),215,Non-crop +216,216,38.23840190779963,1.666609091145675,aasareansah@gmail.com,false,2023-03-07 15:20,15.7 secs,Planet Monthly Mosaics,,POINT(38.23840190779963 1.666609091145675),216,Non-crop +217,217,38.09419036742432,2.9609319691321545,aasareansah@gmail.com,false,2023-03-07 15:20,8.8 secs,Planet Monthly Mosaics,,POINT(38.09419036742432 2.960931969132154),217,Non-crop +218,218,40.11268741129131,0.07066709430285514,aasareansah@gmail.com,false,2023-03-07 15:20,8.0 secs,Planet Monthly Mosaics,,POINT(40.11268741129131 0.070667094302855),218,Non-crop +219,219,35.33297997639196,-0.8617340640007177,aasareansah@gmail.com,false,2023-03-07 15:20,31.1 secs,Planet Monthly Mosaics,,POINT(35.33297997639196 -0.861734064000718),219,Crop +220,220,35.85807814705425,4.10000582662628,aasareansah@gmail.com,false,2023-03-07 15:21,9.3 secs,Planet Monthly Mosaics,,POINT(35.85807814705425 4.10000582662628),220,Non-crop +221,221,34.5729975776531,-0.6466759300640075,aasareansah@gmail.com,false,2023-03-07 15:21,12.1 secs,Planet Monthly Mosaics,,POINT(34.5729975776531 -0.646675930064008),221,Crop +222,222,40.546244593875265,1.1821031768260064,aasareansah@gmail.com,false,2023-03-07 15:21,9.8 secs,Planet Monthly Mosaics,,POINT(40.546244593875265 1.182103176826006),222,Non-crop +223,223,37.548413950439276,-0.41160735763497225,aasareansah@gmail.com,false,2023-03-07 15:21,32.9 secs,Planet Monthly Mosaics,,POINT(37.548413950439276 -0.411607357634972),223,Non-crop +224,224,35.75968549281819,-0.9621064827168123,acgins@umd.edu,false,2023-03-07 21:09,26.7 secs,Planet Monthly Mosaics,,POINT(35.75968549281819 -0.962106482716812),224,Crop +225,225,35.074744972874576,-0.7925917814780629,aasareansah@gmail.com,false,2023-03-07 15:24,43.2 secs,Planet Monthly Mosaics,,POINT(35.074744972874576 -0.792591781478063),225,Crop +226,226,34.46024019008391,3.53618931300853,aasareansah@gmail.com,false,2023-03-07 15:24,9.3 secs,Planet Monthly Mosaics,,POINT(34.46024019008391 3.53618931300853),226,Non-crop +227,227,38.83677570935286,2.824285402842024,aasareansah@gmail.com,false,2023-03-07 15:24,9.5 secs,Planet Monthly Mosaics,,POINT(38.83677570935286 2.824285402842024),227,Non-crop +228,228,37.397368184916004,-0.6140394095524093,aasareansah@gmail.com,false,2023-03-07 15:24,12.2 secs,Planet Monthly Mosaics,,POINT(37.397368184916004 -0.614039409552409),228,Crop +229,229,35.762616150651965,-1.8264944601622872,aasareansah@gmail.com,false,2023-03-07 15:25,9.7 secs,Planet Monthly Mosaics,,POINT(35.762616150651965 -1.826494460162287),229,Non-crop +230,230,36.54848043845705,-1.000772776636237,acgins@umd.edu,false,2023-03-07 21:12,146.9 secs,Planet Monthly Mosaics,,POINT(36.54848043845705 -1.000772776636237),230,Crop +231,231,36.02608988189172,3.627360664812145,aasareansah@gmail.com,false,2023-03-07 15:25,17.9 secs,Planet Monthly Mosaics,,POINT(36.02608988189172 3.627360664812145),231,Non-crop +232,232,33.98152361411099,-0.47111917194585895,aasareansah@gmail.com,false,2023-03-07 15:26,13.4 secs,Planet Monthly Mosaics,,POINT(33.98152361411099 -0.471119171945859),232,Non-crop +233,233,36.08813405957186,-0.941069865990557,acgins@umd.edu,false,2023-03-07 21:13,60.5 secs,Mapbox Satellite,,POINT(36.08813405957186 -0.941069865990557),233,Crop +234,234,37.60598532487981,-0.717185806924993,aasareansah@gmail.com,false,2023-03-07 15:26,8.9 secs,Planet Monthly Mosaics,,POINT(37.60598532487981 -0.717185806924993),234,Non-crop +235,235,34.46747779967232,0.8305382197080086,aasareansah@gmail.com,false,2023-03-07 15:27,39.3 secs,Planet Monthly Mosaics,,POINT(34.46747779967232 0.830538219708009),235,Crop +236,236,34.840773987663475,-0.36183332301270066,aasareansah@gmail.com,false,2023-03-07 15:27,13.4 secs,Planet Monthly Mosaics,,POINT(34.840773987663475 -0.361833323012701),236,Non-crop +237,237,37.41092767037453,-2.625588354308176,aasareansah@gmail.com,false,2023-03-07 15:27,8.9 secs,Planet Monthly Mosaics,,POINT(37.41092767037453 -2.625588354308176),237,Non-crop +238,238,34.5067667042273,4.001073110839722,aasareansah@gmail.com,false,2023-03-07 15:27,9.8 secs,Planet Monthly Mosaics,,POINT(34.5067667042273 4.001073110839722),238,Non-crop +239,239,37.67487855659351,-0.49020904050426295,acgins@umd.edu,false,2023-03-07 21:13,29.6 secs,Planet Monthly Mosaics,,POINT(37.67487855659351 -0.490209040504263),239,Crop +240,240,36.81888775078636,1.9849372691074236,aasareansah@gmail.com,false,2023-03-07 15:27,10.9 secs,Planet Monthly Mosaics,,POINT(36.81888775078636 1.984937269107424),240,Non-crop +241,241,35.24994048248167,-1.0513444774351794,acgins@umd.edu,false,2023-03-07 21:15,114.3 secs,Mapbox Satellite,,POINT(35.24994048248167 -1.051344477435179),241,Crop +242,242,36.38636671973139,-0.5930179628924186,aasareansah@gmail.com,false,2023-03-07 15:28,12.6 secs,Planet Monthly Mosaics,,POINT(36.38636671973139 -0.593017962892419),242,Non-crop +243,243,40.907140441399896,0.6303587257539014,aasareansah@gmail.com,false,2023-03-07 15:28,8.3 secs,Planet Monthly Mosaics,,POINT(40.907140441399896 0.630358725753901),243,Non-crop +244,244,34.96987624938218,1.1655669652004106,aasareansah@gmail.com,false,2023-03-07 15:28,12.3 secs,Planet Monthly Mosaics,,POINT(34.96987624938218 1.165566965200411),244,Crop +245,245,35.35634367392324,1.4582707365296375,aasareansah@gmail.com,false,2023-03-07 15:28,8.7 secs,Planet Monthly Mosaics,,POINT(35.35634367392324 1.458270736529638),245,Non-crop +246,246,40.546097997935725,-0.07612710307294997,aasareansah@gmail.com,false,2023-03-07 15:28,8.8 secs,Planet Monthly Mosaics,,POINT(40.546097997935725 -0.07612710307295),246,Non-crop +247,247,37.16390525162651,-0.7420742630040361,aasareansah@gmail.com,false,2023-03-07 15:28,21.7 secs,Planet Monthly Mosaics,,POINT(37.16390525162651 -0.742074263004036),247,Crop +248,248,34.45068841495048,0.2338980943932806,aasareansah@gmail.com,false,2023-03-07 15:29,29.1 secs,Planet Monthly Mosaics,,POINT(34.45068841495048 0.233898094393281),248,Crop +249,249,37.498286295079225,0.9862352131017155,aasareansah@gmail.com,false,2023-03-07 15:29,16.8 secs,Planet Monthly Mosaics,,POINT(37.498286295079225 0.986235213101716),249,Non-crop +250,250,34.73415295793003,0.018438781875223693,aasareansah@gmail.com,false,2023-03-07 15:29,9.2 secs,Planet Monthly Mosaics,,POINT(34.73415295793003 0.018438781875224),250,Non-crop +251,251,36.85858969388467,-0.4686485504993926,aasareansah@gmail.com,false,2023-03-07 15:30,19.6 secs,Planet Monthly Mosaics,,POINT(36.85858969388467 -0.468648550499393),251,Crop +252,252,35.26890826320225,-0.6564145540663221,aasareansah@gmail.com,false,2023-03-07 15:30,34.1 secs,Planet Monthly Mosaics,,POINT(35.26890826320225 -0.656414554066322),252,Crop +253,253,40.525409290323495,2.6319993375838955,aasareansah@gmail.com,false,2023-03-07 15:30,9.5 secs,Planet Monthly Mosaics,,POINT(40.525409290323495 2.631999337583896),253,Non-crop +254,254,34.92265893447306,0.5353286427591524,acgins@umd.edu,true,2023-03-07 21:29,19.8 secs,,,POINT(34.92265893447306 0.535328642759152),254, +255,255,36.90455311386702,1.1977744701246156,aasareansah@gmail.com,false,2023-03-07 15:32,21.4 secs,Planet Monthly Mosaics,,POINT(36.90455311386702 1.197774470124616),255,Non-crop +256,256,34.5180586260015,0.7789264000463355,aasareansah@gmail.com,false,2023-03-07 15:32,14.3 secs,Planet Monthly Mosaics,,POINT(34.5180586260015 0.778926400046336),256,Non-crop +257,257,36.058209274389014,-1.3463982887488246,aasareansah@gmail.com,false,2023-03-07 15:32,7.8 secs,Planet Monthly Mosaics,,POINT(36.058209274389014 -1.346398288748825),257,Non-crop +258,258,39.395468711706776,2.0383453012130155,aasareansah@gmail.com,false,2023-03-07 15:32,8.6 secs,Planet Monthly Mosaics,,POINT(39.395468711706776 2.038345301213016),258,Non-crop +259,259,39.05025716334389,1.2727642254608016,aasareansah@gmail.com,false,2023-03-07 15:32,11.6 secs,Planet Monthly Mosaics,,POINT(39.05025716334389 1.272764225460802),259,Non-crop +260,260,40.78878641079977,0.947351946589384,aasareansah@gmail.com,false,2023-03-07 15:33,28.9 secs,Planet Monthly Mosaics,,POINT(40.78878641079977 0.947351946589384),260,Non-crop +261,261,39.46290672282505,-0.013057674497699916,aasareansah@gmail.com,false,2023-03-07 15:33,25.7 secs,Planet Monthly Mosaics,,POINT(39.46290672282505 -0.0130576744977),261,Non-crop +262,262,35.19685108842834,3.9040331699588258,aasareansah@gmail.com,false,2023-03-07 15:33,10.8 secs,Planet Monthly Mosaics,,POINT(35.19685108842834 3.904033169958826),262,Non-crop +263,263,37.299210366889746,-0.7882590697784982,aasareansah@gmail.com,false,2023-03-07 15:33,13.5 secs,Planet Monthly Mosaics,,POINT(37.299210366889746 -0.788259069778498),263,Non-crop +264,264,35.134539162750926,1.0904367967402915,aasareansah@gmail.com,false,2023-03-07 15:34,23.7 secs,Planet Monthly Mosaics,,POINT(35.134539162750926 1.090436796740292),264,Crop +265,265,38.19300016102455,-1.6485412210153514,acgins@umd.edu,false,2023-03-07 21:30,42.0 secs,Planet Monthly Mosaics,,POINT(38.19300016102455 -1.648541221015351),265,Crop +266,266,37.73860020267111,-3.331557513616808,aasareansah@gmail.com,false,2023-03-07 15:35,14.3 secs,Planet Monthly Mosaics,,POINT(37.73860020267111 -3.331557513616808),266,Non-crop +267,267,36.183154340115976,2.7001405183904446,aasareansah@gmail.com,false,2023-03-07 15:35,9.1 secs,Planet Monthly Mosaics,,POINT(36.183154340115976 2.700140518390445),267,Non-crop +268,268,39.9003149820409,3.4102972857651417,aasareansah@gmail.com,false,2023-03-07 15:35,8.6 secs,Planet Monthly Mosaics,,POINT(39.9003149820409 3.410297285765142),268,Non-crop +269,269,36.89256033951802,-1.198551976200557,aasareansah@gmail.com,false,2023-03-07 15:35,15.8 secs,Planet Monthly Mosaics,,POINT(36.89256033951802 -1.198551976200557),269,Non-crop +270,270,34.111742992464094,-0.03294774771910352,aasareansah@gmail.com,false,2023-03-07 15:35,12.2 secs,Planet Monthly Mosaics,,POINT(34.111742992464094 -0.032947747719104),270,Non-crop +271,271,38.55271191716017,-1.5870930361786544,aasareansah@gmail.com,false,2023-03-07 15:35,8.5 secs,Planet Monthly Mosaics,,POINT(38.55271191716017 -1.587093036178654),271,Non-crop +272,272,35.79147363846596,1.8930771879554555,aasareansah@gmail.com,false,2023-03-07 15:35,12.1 secs,Planet Monthly Mosaics,,POINT(35.79147363846596 1.893077187955456),272,Non-crop +273,273,35.0802486455078,0.7325502982864748,aasareansah@gmail.com,false,2023-03-07 15:36,12.9 secs,Planet Monthly Mosaics,,POINT(35.0802486455078 0.732550298286475),273,Crop +274,274,36.66787862075995,-0.3370048147857069,aasareansah@gmail.com,false,2023-03-07 15:37,68.4 secs,Planet Monthly Mosaics,,POINT(36.66787862075995 -0.337004814785707),274,Non-crop +275,275,35.382128847748994,-0.8625910503550166,aasareansah@gmail.com,false,2023-03-07 15:37,40.8 secs,Planet Monthly Mosaics,,POINT(35.382128847748994 -0.862591050355017),275,Crop +276,276,37.34840090595225,0.23193127521008539,aasareansah@gmail.com,false,2023-03-07 15:38,25.5 secs,Planet Monthly Mosaics,,POINT(37.34840090595225 0.231931275210085),276,Non-crop +277,277,37.52952524833004,-2.724401129435582,aasareansah@gmail.com,false,2023-03-07 15:39,80.5 secs,Planet Monthly Mosaics,,POINT(37.52952524833004 -2.724401129435582),277,Crop +278,278,34.30386139976155,0.2951159530038938,aasareansah@gmail.com,false,2023-03-07 15:39,16.3 secs,Mapbox Satellite,,POINT(34.30386139976155 0.295115953003894),278,Non-crop +279,279,35.664840363056655,1.690372204559844,aasareansah@gmail.com,false,2023-03-07 15:40,24.8 secs,Planet Monthly Mosaics,,POINT(35.664840363056655 1.690372204559844),279,Non-crop +280,280,39.19739507190026,-1.2216252242734178,aasareansah@gmail.com,false,2023-03-07 15:40,8.0 secs,Planet Monthly Mosaics,,POINT(39.19739507190026 -1.221625224273418),280,Non-crop +281,281,35.184928340727154,1.8744824920807754,aasareansah@gmail.com,false,2023-03-07 15:40,23.3 secs,Planet Monthly Mosaics,,POINT(35.184928340727154 1.874482492080775),281,Non-crop +282,282,37.78857700274938,-3.3771286904231905,aasareansah@gmail.com,false,2023-03-07 15:40,9.2 secs,Planet Monthly Mosaics,,POINT(37.78857700274938 -3.37712869042319),282,Non-crop +283,283,36.68502729602952,2.514325852545885,aasareansah@gmail.com,false,2023-03-07 15:40,10.1 secs,Planet Monthly Mosaics,,POINT(36.68502729602952 2.514325852545885),283,Non-crop +284,284,35.237040051019214,1.6524385569289506,acgins@umd.edu,false,2023-03-07 21:31,64.2 secs,Planet Monthly Mosaics,,POINT(35.237040051019214 1.652438556928951),284,Crop +285,285,36.53526300735865,-0.4683613786490718,acgins@umd.edu,false,2023-03-07 21:32,31.9 secs,Planet Monthly Mosaics,,POINT(36.53526300735865 -0.468361378649072),285,Crop +286,286,38.0760525856689,-0.11913545569567449,aasareansah@gmail.com,false,2023-03-07 15:41,19.8 secs,Planet Monthly Mosaics,,POINT(38.0760525856689 -0.119135455695674),286,Non-crop +287,287,34.35644028067466,0.5608866070711257,aasareansah@gmail.com,false,2023-03-07 15:42,15.1 secs,Planet Monthly Mosaics,,POINT(34.35644028067466 0.560886607071126),287,Crop +288,288,38.65484794845974,-0.4860956688299213,aasareansah@gmail.com,false,2023-03-07 15:42,9.7 secs,Planet Monthly Mosaics,,POINT(38.65484794845974 -0.486095668829921),288,Non-crop +289,289,35.93734608846335,-1.057140931452576,aasareansah@gmail.com,false,2023-03-07 15:42,12.9 secs,Planet Monthly Mosaics,,POINT(35.93734608846335 -1.057140931452576),289,Non-crop +290,290,37.80582391504641,3.8350499412603556,aasareansah@gmail.com,false,2023-03-07 15:42,9.2 secs,Planet Monthly Mosaics,,POINT(37.80582391504641 3.835049941260356),290,Non-crop +291,291,35.25737937467433,-0.8666069188032641,aasareansah@gmail.com,false,2023-03-07 15:42,17.7 secs,Planet Monthly Mosaics,,POINT(35.25737937467433 -0.866606918803264),291,Non-crop +292,292,34.35788383131346,0.19694288433069554,acgins@umd.edu,false,2023-03-07 21:33,78.5 secs,Sentinel-2,,POINT(34.35788383131346 0.196942884330696),292,Crop +293,293,34.818940898978965,-1.2688823327413155,aasareansah@gmail.com,false,2023-03-07 15:43,13.5 secs,Planet Monthly Mosaics,,POINT(34.818940898978965 -1.268882332741316),293,Non-crop +294,294,34.93880589858098,2.3229353279690117,aasareansah@gmail.com,false,2023-03-07 15:43,28.6 secs,Planet Monthly Mosaics,,POINT(34.93880589858098 2.322935327969012),294,Non-crop +295,295,39.054542526331986,0.0026259847551115003,aasareansah@gmail.com,false,2023-03-07 15:43,9.6 secs,Planet Monthly Mosaics,,POINT(39.054542526331986 0.002625984755112),295,Non-crop +296,296,36.07763478509327,-1.1708826042522547,acgins@umd.edu,false,2023-03-07 21:33,31.7 secs,Planet Monthly Mosaics,,POINT(36.07763478509327 -1.170882604252255),296,Crop +297,297,34.90669887109912,0.7293370301766702,aasareansah@gmail.com,false,2023-03-07 15:44,30.7 secs,Planet Monthly Mosaics,,POINT(34.90669887109912 0.72933703017667),297,Crop +298,298,38.79853398079912,0.9642160077813193,aasareansah@gmail.com,false,2023-03-07 15:45,8.8 secs,Planet Monthly Mosaics,,POINT(38.79853398079912 0.964216007781319),298,Non-crop +299,299,37.553857774558885,-2.9026633735747365,aasareansah@gmail.com,false,2023-03-07 15:45,38.9 secs,Planet Monthly Mosaics,,POINT(37.553857774558885 -2.902663373574736),299,Non-crop +300,300,38.01347608722678,2.900570657102886,aasareansah@gmail.com,false,2023-03-07 15:45,11.4 secs,Planet Monthly Mosaics,,POINT(38.01347608722678 2.900570657102886),300,Non-crop +301,301,37.132692772610476,0.39776039039835437,aasareansah@gmail.com,false,2023-03-07 15:45,8.9 secs,Planet Monthly Mosaics,,POINT(37.132692772610476 0.397760390398354),301,Non-crop +302,302,37.9392642060917,2.9892719733950126,aasareansah@gmail.com,false,2023-03-07 15:46,8.7 secs,Planet Monthly Mosaics,,POINT(37.9392642060917 2.989271973395013),302,Non-crop +303,303,34.51506919888127,0.1202453794900851,aasareansah@gmail.com,false,2023-03-07 15:46,14.8 secs,Planet Monthly Mosaics,,POINT(34.51506919888127 0.120245379490085),303,Crop +304,304,39.63767229599631,-0.12049561541535113,aasareansah@gmail.com,false,2023-03-07 15:46,8.8 secs,Planet Monthly Mosaics,,POINT(39.63767229599631 -0.120495615415351),304,Non-crop +305,305,40.485326212115005,3.093570121762987,aasareansah@gmail.com,false,2023-03-07 15:46,8.2 secs,Planet Monthly Mosaics,,POINT(40.485326212115005 3.093570121762987),305,Non-crop +306,306,36.85239673523007,-0.05853659594733882,aasareansah@gmail.com,false,2023-03-07 15:46,9.8 secs,Planet Monthly Mosaics,,POINT(36.85239673523007 -0.058536595947339),306,Non-crop +307,307,37.5763483597854,-0.9823710756791123,aasareansah@gmail.com,false,2023-03-07 15:47,31.9 secs,Planet Monthly Mosaics,,POINT(37.5763483597854 -0.982371075679112),307,Non-crop +308,308,39.55791935681194,1.5349026436085218,aasareansah@gmail.com,false,2023-03-07 15:47,9.8 secs,Planet Monthly Mosaics,,POINT(39.55791935681194 1.534902643608522),308,Non-crop +309,309,37.699949750619936,0.811948379073906,aasareansah@gmail.com,false,2023-03-07 15:47,8.5 secs,Planet Monthly Mosaics,,POINT(37.699949750619936 0.811948379073906),309,Non-crop +310,310,37.61422072618653,-2.0872413462062473,acgins@umd.edu,false,2023-03-07 21:37,230.8 secs,Sentinel-2,,POINT(37.61422072618653 -2.087241346206247),310,Crop +311,311,41.07513171921359,3.1888676288992936,aasareansah@gmail.com,false,2023-03-07 15:47,14.5 secs,Planet Monthly Mosaics,,POINT(41.07513171921359 3.188867628899294),311,Non-crop +312,312,38.15815408847578,-0.8111453505853832,aasareansah@gmail.com,false,2023-03-07 15:47,9.0 secs,Planet Monthly Mosaics,,POINT(38.15815408847578 -0.811145350585383),312,Non-crop +313,313,34.736270179275344,-0.5624127830726717,aasareansah@gmail.com,false,2023-03-07 15:48,19.3 secs,Planet Monthly Mosaics,,POINT(34.736270179275344 -0.562412783072672),313,Crop +314,314,35.375810528396606,-0.21120699908951038,aasareansah@gmail.com,false,2023-03-07 15:48,13.3 secs,Planet Monthly Mosaics,,POINT(35.375810528396606 -0.21120699908951),314,Crop +315,315,37.21904951713753,-0.7068700226033653,aasareansah@gmail.com,false,2023-03-07 15:48,11.7 secs,Planet Monthly Mosaics,,POINT(37.21904951713753 -0.706870022603365),315,Non-crop +316,316,39.83400661063378,-1.68890740390545,aasareansah@gmail.com,false,2023-03-07 15:48,10.5 secs,Planet Monthly Mosaics,,POINT(39.83400661063378 -1.68890740390545),316,Non-crop +317,317,34.191058880953285,-0.9215307463490227,acgins@umd.edu,false,2023-03-07 21:38,39.2 secs,Planet Monthly Mosaics,,POINT(34.191058880953285 -0.921530746349023),317,Crop +318,318,34.49184864640101,-0.7460763000252552,acgins@umd.edu,false,2023-03-07 21:39,84.9 secs,Mapbox Satellite,,POINT(34.49184864640101 -0.746076300025255),318,Crop +319,319,37.651127920351016,-0.6203277536564297,aasareansah@gmail.com,false,2023-03-07 15:49,12.9 secs,Planet Monthly Mosaics,,POINT(37.651127920351016 -0.62032775365643),319,Non-crop +320,320,38.75780790144454,-0.6667358112308936,aasareansah@gmail.com,false,2023-03-07 15:49,8.3 secs,Planet Monthly Mosaics,,POINT(38.75780790144454 -0.666735811230894),320,Non-crop +321,321,34.94783457049091,-0.11246745761080304,aasareansah@gmail.com,false,2023-03-07 15:49,12.4 secs,Planet Monthly Mosaics,,POINT(34.94783457049091 -0.112467457610803),321,Non-crop +322,322,39.229061071429356,-0.666375893413772,aasareansah@gmail.com,false,2023-03-07 15:49,8.0 secs,Planet Monthly Mosaics,,POINT(39.229061071429356 -0.666375893413772),322,Non-crop +323,323,38.88788044369188,-4.24884778504972,aasareansah@gmail.com,false,2023-03-07 15:49,8.4 secs,Planet Monthly Mosaics,,POINT(38.88788044369188 -4.24884778504972),323,Non-crop +324,324,36.71627653401663,2.0002891318295455,aasareansah@gmail.com,false,2023-03-07 15:50,9.6 secs,Planet Monthly Mosaics,,POINT(36.71627653401663 2.000289131829546),324,Non-crop +325,325,35.99051353043517,-1.0041755970886126,aasareansah@gmail.com,false,2023-03-07 15:50,18.8 secs,Planet Monthly Mosaics,,POINT(35.99051353043517 -1.004175597088613),325,Non-crop +326,326,34.693116889906136,0.7036994275098215,aasareansah@gmail.com,false,2023-03-07 15:50,16.1 secs,Planet Monthly Mosaics,,POINT(34.693116889906136 0.703699427509822),326,Crop +327,327,36.80110616930354,-1.3991582831341616,aasareansah@gmail.com,false,2023-03-07 15:50,13.0 secs,Planet Monthly Mosaics,,POINT(36.80110616930354 -1.399158283134162),327,Non-crop +328,328,34.359713263958206,3.7495077872591676,aasareansah@gmail.com,false,2023-03-07 15:50,9.7 secs,Planet Monthly Mosaics,,POINT(34.359713263958206 3.749507787259168),328,Non-crop +329,329,34.89872798832474,3.7711850926712276,aasareansah@gmail.com,false,2023-03-07 15:50,9.3 secs,Planet Monthly Mosaics,,POINT(34.89872798832474 3.771185092671228),329,Non-crop +330,330,34.75823814690865,0.768390378516022,acgins@umd.edu,false,2023-03-07 21:41,72.9 secs,Mapbox Satellite,,POINT(34.75823814690865 0.768390378516022),330,Crop +331,331,35.03043307620979,-0.435833843994421,aasareansah@gmail.com,false,2023-03-07 15:51,25.0 secs,Planet Monthly Mosaics,,POINT(35.03043307620979 -0.435833843994421),331,Crop +332,332,39.91725992960783,-2.536230581743613,aasareansah@gmail.com,false,2023-03-07 15:51,12.6 secs,Planet Monthly Mosaics,,POINT(39.91725992960783 -2.536230581743613),332,Non-crop +333,333,38.04110617648035,2.134810668583166,aasareansah@gmail.com,false,2023-03-07 15:51,9.8 secs,Planet Monthly Mosaics,,POINT(38.04110617648035 2.134810668583166),333,Non-crop +334,334,38.616261925366786,2.5167846397007954,aasareansah@gmail.com,false,2023-03-07 15:51,8.8 secs,Planet Monthly Mosaics,,POINT(38.616261925366786 2.516784639700795),334,Non-crop +335,335,36.910665570255915,-1.1618142066887174,aasareansah@gmail.com,false,2023-03-07 15:52,11.6 secs,Planet Monthly Mosaics,,POINT(36.910665570255915 -1.161814206688717),335,Non-crop +336,336,36.42768875867901,2.649630024893417,aasareansah@gmail.com,false,2023-03-07 15:52,10.0 secs,Planet Monthly Mosaics,,POINT(36.42768875867901 2.649630024893417),336,Non-crop +337,337,37.41839115014418,-1.0816838435200997,aasareansah@gmail.com,false,2023-03-07 15:52,36.6 secs,Planet Monthly Mosaics,,POINT(37.41839115014418 -1.0816838435201),337,Non-crop +338,338,37.10693932494706,-0.6350245887261098,aasareansah@gmail.com,false,2023-03-07 15:52,18.5 secs,Planet Monthly Mosaics,,POINT(37.10693932494706 -0.63502458872611),338,Non-crop +339,339,36.984429661995755,3.311675194970429,aasareansah@gmail.com,false,2023-03-07 15:53,8.4 secs,Planet Monthly Mosaics,,POINT(36.984429661995755 3.311675194970429),339,Non-crop +340,340,36.144412583364996,0.004963373055491912,acgins@umd.edu,false,2023-03-07 21:44,214.8 secs,Planet Monthly Mosaics,,POINT(36.144412583364996 0.004963373055492),340,Crop +341,341,36.70078650447418,3.309703935432483,aasareansah@gmail.com,false,2023-03-07 15:53,8.4 secs,Planet Monthly Mosaics,,POINT(36.70078650447418 3.309703935432483),341,Non-crop +342,342,38.75161279448816,2.378729348687549,aasareansah@gmail.com,false,2023-03-07 15:53,8.3 secs,Planet Monthly Mosaics,,POINT(38.75161279448816 2.378729348687549),342,Non-crop +343,343,40.06707636849588,3.45275268077823,aasareansah@gmail.com,false,2023-03-07 15:53,12.2 secs,Planet Monthly Mosaics,,POINT(40.06707636849588 3.45275268077823),343,Non-crop +344,344,36.56210500652624,-1.0837246415363992,aasareansah@gmail.com,false,2023-03-07 15:53,8.7 secs,Planet Monthly Mosaics,,POINT(36.56210500652624 -1.083724641536399),344,Non-crop +345,345,34.336041577902456,0.7166491969812425,acgins@umd.edu,false,2023-03-07 21:46,139.9 secs,Mapbox Satellite,,POINT(34.336041577902456 0.716649196981242),345,Crop +346,346,39.598853154828326,2.8858817878871452,aasareansah@gmail.com,false,2023-03-07 15:54,9.1 secs,Planet Monthly Mosaics,,POINT(39.598853154828326 2.885881787887145),346,Non-crop +347,347,36.43918846011223,1.440504544935712,aasareansah@gmail.com,false,2023-03-07 15:54,8.6 secs,Planet Monthly Mosaics,,POINT(36.43918846011223 1.440504544935712),347,Non-crop +348,348,35.54210939603061,-1.327991441743508,aasareansah@gmail.com,false,2023-03-07 15:54,8.3 secs,Planet Monthly Mosaics,,POINT(35.54210939603061 -1.327991441743508),348,Non-crop +349,349,34.24661018198986,3.7788091335579623,aasareansah@gmail.com,false,2023-03-07 15:54,8.4 secs,Planet Monthly Mosaics,,POINT(34.24661018198986 3.778809133557962),349,Non-crop +350,350,38.155177833382425,-0.9094747900561307,aasareansah@gmail.com,false,2023-03-07 15:54,16.8 secs,Planet Monthly Mosaics,,POINT(38.155177833382425 -0.909474790056131),350,Non-crop +351,351,37.72392326099627,-2.8587133707172647,aasareansah@gmail.com,false,2023-03-07 15:54,8.7 secs,Planet Monthly Mosaics,,POINT(37.72392326099627 -2.858713370717265),351,Non-crop +352,352,39.185711371300165,0.11214360263653626,aasareansah@gmail.com,false,2023-03-07 15:54,8.1 secs,Planet Monthly Mosaics,,POINT(39.185711371300165 0.112143602636536),352,Non-crop +353,353,36.81517663408579,0.06603508799998077,aasareansah@gmail.com,false,2023-03-07 15:54,9.3 secs,Planet Monthly Mosaics,,POINT(36.81517663408579 0.066035087999981),353,Non-crop +354,354,39.73857946305931,1.3005086068260963,aasareansah@gmail.com,false,2023-03-07 15:55,8.1 secs,Planet Monthly Mosaics,,POINT(39.73857946305931 1.300508606826096),354,Non-crop +355,355,34.89094893452659,0.5432189512967784,aasareansah@gmail.com,false,2023-03-07 15:55,29.7 secs,Planet Monthly Mosaics,,POINT(34.89094893452659 0.543218951296778),355,Crop +356,356,35.34730497147133,1.8974096389644381,aasareansah@gmail.com,false,2023-03-07 15:55,13.1 secs,Planet Monthly Mosaics,,POINT(35.34730497147133 1.897409638964438),356,Non-crop +357,357,39.4560023573566,-3.7589481349304537,aasareansah@gmail.com,false,2023-03-07 15:55,13.6 secs,Planet Monthly Mosaics,,POINT(39.4560023573566 -3.758948134930454),357,Non-crop +358,358,34.19309376608378,-0.7112667390382308,acgins@umd.edu,false,2023-03-07 21:48,87.5 secs,Planet Monthly Mosaics,,POINT(34.19309376608378 -0.711266739038231),358,Crop +359,359,38.08520562034121,-0.7104578632257538,aasareansah@gmail.com,false,2023-03-07 15:56,24.5 secs,Planet Monthly Mosaics,,POINT(38.08520562034121 -0.710457863225754),359,Non-crop +360,360,39.8764623690857,-0.41394507470955105,acgins@umd.edu,false,2023-03-07 21:49,65.4 secs,Mapbox Satellite,,POINT(39.8764623690857 -0.413945074709551),360,Non-crop +361,361,36.59090129950348,4.401266892103294,aasareansah@gmail.com,false,2023-03-07 15:57,26.6 secs,Planet Monthly Mosaics,,POINT(36.59090129950348 4.401266892103294),361,Non-crop +362,362,36.80582306681952,3.5421413369810377,aasareansah@gmail.com,false,2023-03-07 15:57,10.2 secs,Planet Monthly Mosaics,,POINT(36.80582306681952 3.542141336981038),362,Non-crop +363,363,40.84664060103932,1.7168775775104577,aasareansah@gmail.com,false,2023-03-07 15:57,11.0 secs,Planet Monthly Mosaics,,POINT(40.84664060103932 1.716877577510458),363,Non-crop +364,364,40.68156500663407,1.129225977950579,aasareansah@gmail.com,false,2023-03-07 15:57,15.2 secs,Planet Monthly Mosaics,,POINT(40.68156500663407 1.129225977950579),364,Non-crop +365,365,38.55642953523764,2.918040794846058,aasareansah@gmail.com,false,2023-03-07 15:57,10.1 secs,Planet Monthly Mosaics,,POINT(38.55642953523764 2.918040794846058),365,Non-crop +366,366,39.0709024919474,-3.7913328870442045,aasareansah@gmail.com,false,2023-03-07 15:57,12.4 secs,Planet Monthly Mosaics,,POINT(39.0709024919474 -3.791332887044204),366,Non-crop +367,367,36.638130724991306,-0.49460489559864695,aasareansah@gmail.com,false,2023-03-07 15:58,12.2 secs,Planet Monthly Mosaics,,POINT(36.638130724991306 -0.494604895598647),367,Non-crop +368,368,37.63079766174867,-0.053604990844186254,aasareansah@gmail.com,false,2023-03-07 15:58,22.7 secs,Planet Monthly Mosaics,,POINT(37.63079766174867 -0.053604990844186),368,Crop +369,369,35.827907745167956,-0.853811982262311,aasareansah@gmail.com,false,2023-03-07 15:58,8.6 secs,Planet Monthly Mosaics,,POINT(35.827907745167956 -0.853811982262311),369,Non-crop +370,370,39.167720267265224,1.9374319528339419,aasareansah@gmail.com,false,2023-03-07 15:58,12.7 secs,Planet Monthly Mosaics,,POINT(39.167720267265224 1.937431952833942),370,Non-crop +371,371,34.212762133967715,-0.8357804134194107,acgins@umd.edu,false,2023-03-07 21:51,101.5 secs,Planet Monthly Mosaics,,POINT(34.212762133967715 -0.835780413419411),371,Crop +372,372,37.96110516185624,0.14781455235668403,aasareansah@gmail.com,false,2023-03-07 15:59,13.9 secs,Planet Monthly Mosaics,,POINT(37.96110516185624 0.147814552356684),372,Non-crop +373,373,34.47463753982868,0.10990526138305981,acgins@umd.edu,false,2023-03-07 21:54,218.7 secs,Planet Monthly Mosaics,,POINT(34.47463753982868 0.10990526138306),373,Non-crop +374,374,39.69251931795742,-2.5541267260917877,aasareansah@gmail.com,false,2023-03-07 15:59,11.7 secs,Planet Monthly Mosaics,,POINT(39.69251931795742 -2.554126726091788),374,Non-crop +375,375,36.68244024793016,-1.8676442073681212,aasareansah@gmail.com,false,2023-03-07 15:59,8.5 secs,Planet Monthly Mosaics,,POINT(36.68244024793016 -1.867644207368121),375,Non-crop +376,376,36.853159792048174,3.4206556695885375,aasareansah@gmail.com,false,2023-03-07 15:59,9.1 secs,Planet Monthly Mosaics,,POINT(36.853159792048174 3.420655669588538),376,Non-crop +377,377,38.92586119879842,1.3194550813766501,aasareansah@gmail.com,false,2023-03-07 16:00,12.7 secs,Planet Monthly Mosaics,,POINT(38.92586119879842 1.31945508137665),377,Non-crop +378,378,38.69999902903868,2.7545836053240813,aasareansah@gmail.com,false,2023-03-07 16:00,8.9 secs,Planet Monthly Mosaics,,POINT(38.69999902903868 2.754583605324081),378,Non-crop +379,379,34.83181317134586,1.144867592720978,aasareansah@gmail.com,false,2023-03-07 16:00,12.2 secs,Planet Monthly Mosaics,,POINT(34.83181317134586 1.144867592720978),379,Crop +380,380,40.90693313748294,1.8439313546702625,aasareansah@gmail.com,false,2023-03-07 16:00,8.2 secs,Planet Monthly Mosaics,,POINT(40.90693313748294 1.843931354670262),380,Non-crop +381,381,34.1304218109236,-0.5631023328537231,aasareansah@gmail.com,false,2023-03-07 16:00,9.0 secs,Planet Monthly Mosaics,,POINT(34.1304218109236 -0.563102332853723),381,Non-crop +382,382,39.200957664948234,-2.5959413740347235,aasareansah@gmail.com,false,2023-03-07 16:00,11.6 secs,Planet Monthly Mosaics,,POINT(39.200957664948234 -2.595941374034724),382,Non-crop +383,383,38.63760081316255,1.1224819906657701,aasareansah@gmail.com,false,2023-03-07 16:00,7.6 secs,Planet Monthly Mosaics,,POINT(38.63760081316255 1.12248199066577),383,Non-crop +384,384,34.795026265560566,0.6743079317838352,acgins@umd.edu,false,2023-03-07 21:55,61.5 secs,Mapbox Satellite,,POINT(34.795026265560566 0.674307931783835),384,Crop +385,385,40.272953061237764,-1.7216502223117192,aasareansah@gmail.com,false,2023-03-07 16:01,8.9 secs,Planet Monthly Mosaics,,POINT(40.272953061237764 -1.721650222311719),385,Non-crop +386,386,34.511216655298945,-0.40840217196105755,aasareansah@gmail.com,false,2023-03-07 16:01,8.5 secs,Planet Monthly Mosaics,,POINT(34.511216655298945 -0.408402171961058),386,Non-crop +387,387,38.532332486665574,-3.4722325425673475,aasareansah@gmail.com,false,2023-03-07 16:01,7.3 secs,Planet Monthly Mosaics,,POINT(38.532332486665574 -3.472232542567348),387,Non-crop +388,388,34.824113704860466,1.2169539748385623,aasareansah@gmail.com,false,2023-03-07 16:01,12.7 secs,Planet Monthly Mosaics,,POINT(34.824113704860466 1.216953974838562),388,Crop +389,389,37.388102475893305,-0.7540557617057277,aasareansah@gmail.com,false,2023-03-07 16:02,29.1 secs,Planet Monthly Mosaics,,POINT(37.388102475893305 -0.754055761705728),389,Non-crop +390,390,37.42891460399352,-0.45483461285550036,acgins@umd.edu,false,2023-03-07 21:58,144.0 secs,Mapbox Satellite,,POINT(37.42891460399352 -0.4548346128555),390,Crop +391,391,36.967887534726245,-0.8095488907087274,aasareansah@gmail.com,false,2023-03-07 16:02,19.7 secs,Planet Monthly Mosaics,,POINT(36.967887534726245 -0.809548890708727),391,Non-crop +392,392,35.527048399962815,4.0989025202231,aasareansah@gmail.com,false,2023-03-07 16:02,10.0 secs,Planet Monthly Mosaics,,POINT(35.527048399962815 4.0989025202231),392,Non-crop +393,393,36.924922755405284,-1.1232711340297845,aasareansah@gmail.com,false,2023-03-07 16:02,9.2 secs,Planet Monthly Mosaics,,POINT(36.924922755405284 -1.123271134029784),393,Non-crop +394,394,37.257050079664545,0.7989869091198432,aasareansah@gmail.com,false,2023-03-07 16:02,8.5 secs,Planet Monthly Mosaics,,POINT(37.257050079664545 0.798986909119843),394,Non-crop +395,395,38.95454874266865,3.252171003985754,aasareansah@gmail.com,false,2023-03-07 16:02,9.1 secs,Planet Monthly Mosaics,,POINT(38.95454874266865 3.252171003985754),395,Non-crop +396,396,35.341565025437944,-0.29359632555640297,acgins@umd.edu,true,2023-03-07 22:04,397.8 secs,,,POINT(35.341565025437944 -0.293596325556403),396, +397,397,35.79074795959633,-1.9464192254714852,aasareansah@gmail.com,false,2023-03-07 16:03,12.1 secs,Planet Monthly Mosaics,,POINT(35.79074795959633 -1.946419225471485),397,Non-crop +398,398,34.651917881792194,0.5319953940694888,aasareansah@gmail.com,false,2023-03-07 16:03,8.1 secs,Planet Monthly Mosaics,,POINT(34.651917881792194 0.531995394069489),398,Non-crop +399,399,41.1306620114978,3.1745220367098654,aasareansah@gmail.com,false,2023-03-07 16:03,7.8 secs,Planet Monthly Mosaics,,POINT(41.1306620114978 3.174522036709865),399,Non-crop +400,400,34.4371951436428,-0.7709979914980767,aasareansah@gmail.com,false,2023-03-07 16:03,18.7 secs,Planet Monthly Mosaics,,POINT(34.4371951436428 -0.770997991498077),400,Crop +401,401,34.558043775685306,3.125264995688962,aasareansah@gmail.com,false,2023-03-07 16:03,9.4 secs,Planet Monthly Mosaics,,POINT(34.558043775685306 3.125264995688962),401,Non-crop +402,402,37.981040475396064,3.6028567914524987,aasareansah@gmail.com,false,2023-03-07 16:04,22.9 secs,Planet Monthly Mosaics,,POINT(37.981040475396064 3.602856791452499),402,Non-crop +403,403,38.14269055562707,0.6772356097837042,aasareansah@gmail.com,false,2023-03-07 16:04,7.9 secs,Planet Monthly Mosaics,,POINT(38.14269055562707 0.677235609783704),403,Non-crop +404,404,34.46795120792515,0.6249513233906141,acgins@umd.edu,false,2023-03-07 22:07,139.8 secs,Sentinel-2,,POINT(34.46795120792515 0.624951323390614),404,Crop +405,405,35.18905016180328,-0.9303939864286976,aasareansah@gmail.com,false,2023-03-07 16:05,41.3 secs,Planet Monthly Mosaics,,POINT(35.18905016180328 -0.930393986428698),405,Crop +406,406,34.680031744225815,-1.056354575251415,acgins@umd.edu,false,2023-03-07 22:10,189.6 secs,Mapbox Satellite,,POINT(34.680031744225815 -1.056354575251415),406,Non-crop +407,407,38.584107662616205,-4.000071152288873,aasareansah@gmail.com,false,2023-03-07 16:05,10.0 secs,Planet Monthly Mosaics,,POINT(38.584107662616205 -4.000071152288873),407,Non-crop +408,408,34.54922635792607,0.568676232467724,aasareansah@gmail.com,false,2023-03-07 16:05,12.9 secs,Planet Monthly Mosaics,,POINT(34.54922635792607 0.568676232467724),408,Non-crop +409,409,34.46976620330974,0.5891906542491239,acgins@umd.edu,false,2023-03-07 22:11,43.5 secs,Mapbox Satellite,,POINT(34.46976620330974 0.589190654249124),409,Crop +410,410,34.317026783547334,-0.14004161925952469,aasareansah@gmail.com,false,2023-03-07 16:06,24.2 secs,Planet Monthly Mosaics,,POINT(34.317026783547334 -0.140041619259525),410,Crop +411,411,37.67958111982131,-0.031077482460660336,aasareansah@gmail.com,false,2023-03-07 16:06,12.3 secs,Planet Monthly Mosaics,,POINT(37.67958111982131 -0.03107748246066),411,Non-crop +412,412,34.97698104242654,0.5048837047733515,acgins@umd.edu,false,2023-03-07 22:13,114.7 secs,Planet Monthly Mosaics,,POINT(34.97698104242654 0.504883704773352),412,Crop +413,413,39.42808479465304,2.647284919989776,aasareansah@gmail.com,false,2023-03-07 16:06,11.4 secs,Planet Monthly Mosaics,,POINT(39.42808479465304 2.647284919989776),413,Non-crop +414,414,36.941839061604995,3.049551867540078,aasareansah@gmail.com,false,2023-03-07 16:06,7.8 secs,Planet Monthly Mosaics,,POINT(36.941839061604995 3.049551867540078),414,Non-crop +415,415,34.79602405819177,0.09970469632021167,aasareansah@gmail.com,false,2023-03-07 16:06,21.4 secs,Planet Monthly Mosaics,,POINT(34.79602405819177 0.099704696320212),415,Non-crop +416,416,34.566268622694864,-0.9309187229149422,aasareansah@gmail.com,false,2023-03-07 16:07,23.7 secs,Planet Monthly Mosaics,,POINT(34.566268622694864 -0.930918722914942),416,Crop +417,417,34.193493927332696,-0.7542905932579258,aasareansah@gmail.com,false,2023-03-07 16:07,15.2 secs,Planet Monthly Mosaics,,POINT(34.193493927332696 -0.754290593257926),417,Crop +418,418,37.15096183503388,-0.33350560480048325,acgins@umd.edu,false,2023-03-07 22:14,63.1 secs,Mapbox Satellite,,POINT(37.15096183503388 -0.333505604800483),418,Crop +419,419,34.51308490339396,-0.7547247822555792,acgins@umd.edu,false,2023-03-08 01:25,11512.2 secs,Mapbox Satellite,,POINT(34.51308490339396 -0.754724782255579),419,Non-crop +420,420,36.50791814531247,-0.7870475476290038,adadebay@umd.edu,false,2023-03-08 17:58,28.5 secs,Mapbox Satellite,,POINT(36.50791814531247 -0.787047547629004),420,Crop +421,421,37.82868384399238,3.714236460960642,aasareansah@gmail.com,false,2023-03-07 16:08,9.2 secs,Planet Monthly Mosaics,,POINT(37.82868384399238 3.714236460960642),421,Non-crop +422,422,40.009040766686724,-0.3419198637588498,aasareansah@gmail.com,false,2023-03-07 16:08,9.5 secs,Planet Monthly Mosaics,,POINT(40.009040766686724 -0.34191986375885),422,Non-crop +423,423,34.491276416658955,-1.2346310142928727,adadebay@umd.edu,false,2023-03-08 17:58,11.5 secs,Mapbox Satellite,,POINT(34.491276416658955 -1.234631014292873),423,Crop +424,424,35.15441964409439,-0.43733764965412875,aasareansah@gmail.com,false,2023-03-07 16:12,161.7 secs,Mapbox Satellite,,POINT(35.15441964409439 -0.437337649654129),424,Non-crop +425,425,35.28686492785766,4.213669588373389,aasareansah@gmail.com,false,2023-03-07 16:13,62.7 secs,Planet Monthly Mosaics,,POINT(35.28686492785766 4.213669588373389),425,Non-crop +426,426,40.499705577963645,3.056516667636345,aasareansah@gmail.com,false,2023-03-07 16:13,10.6 secs,Planet Monthly Mosaics,,POINT(40.499705577963645 3.056516667636345),426,Non-crop +427,427,40.30464352437439,-0.4326296899359431,aasareansah@gmail.com,false,2023-03-07 16:13,10.4 secs,Planet Monthly Mosaics,,POINT(40.30464352437439 -0.432629689935943),427,Non-crop +428,428,39.34917441922158,1.099228413243411,aasareansah@gmail.com,false,2023-03-07 16:13,8.1 secs,Planet Monthly Mosaics,,POINT(39.34917441922158 1.099228413243411),428,Non-crop +429,429,34.26078466118539,-1.0247476972595801,adadebay@umd.edu,false,2023-03-08 17:59,10.5 secs,Mapbox Satellite,,POINT(34.26078466118539 -1.02474769725958),429,Crop +430,430,39.74078659873234,-1.1870999208832935,aasareansah@gmail.com,false,2023-03-07 16:13,8.1 secs,Planet Monthly Mosaics,,POINT(39.74078659873234 -1.187099920883294),430,Non-crop +431,431,34.76303145495171,0.4669028149604851,aasareansah@gmail.com,false,2023-03-07 16:14,34.1 secs,Planet Monthly Mosaics,,POINT(34.76303145495171 0.466902814960485),431,Crop +432,432,37.39470110843973,-0.5501557583050474,aasareansah@gmail.com,false,2023-03-07 16:14,17.3 secs,Planet Monthly Mosaics,,POINT(37.39470110843973 -0.550155758305047),432,Non-crop +433,433,37.174734166271435,-2.02859100605471,aasareansah@gmail.com,false,2023-03-07 16:14,9.2 secs,Planet Monthly Mosaics,,POINT(37.174734166271435 -2.02859100605471),433,Non-crop +434,434,35.7713835612932,1.7742478484709758,aasareansah@gmail.com,false,2023-03-07 16:14,11.7 secs,Planet Monthly Mosaics,,POINT(35.7713835612932 1.774247848470976),434,Non-crop +435,435,35.054458324171065,-0.44070243296815076,aasareansah@gmail.com,false,2023-03-07 16:14,17.2 secs,Planet Monthly Mosaics,,POINT(35.054458324171065 -0.440702432968151),435,Crop +436,436,35.535317107443134,4.0300216964176565,aasareansah@gmail.com,false,2023-03-07 16:15,9.3 secs,Planet Monthly Mosaics,,POINT(35.535317107443134 4.030021696417656),436,Non-crop +437,437,36.195615641572736,1.2500737632246104,aasareansah@gmail.com,false,2023-03-07 16:15,8.9 secs,Planet Monthly Mosaics,,POINT(36.195615641572736 1.25007376322461),437,Non-crop +438,438,34.585500564173465,-1.1006159436375704,aasareansah@gmail.com,false,2023-03-07 16:15,23.8 secs,Planet Monthly Mosaics,,POINT(34.585500564173465 -1.10061594363757),438,Crop +439,439,36.93681116460859,-0.8568247999425626,aasareansah@gmail.com,false,2023-03-07 16:15,7.5 secs,Planet Monthly Mosaics,,POINT(36.93681116460859 -0.856824799942563),439,Non-crop +440,440,34.653940619977,-1.2979980707441048,aasareansah@gmail.com,false,2023-03-07 16:15,8.3 secs,Planet Monthly Mosaics,,POINT(34.653940619977 -1.297998070744105),440,Non-crop +441,441,39.95325288909631,-3.0539655940358257,aasareansah@gmail.com,false,2023-03-07 16:15,10.1 secs,Planet Monthly Mosaics,,POINT(39.95325288909631 -3.053965594035826),441,Non-crop +442,442,34.79643269660636,0.954060045767657,adadebay@umd.edu,false,2023-03-08 17:59,44.1 secs,Mapbox Satellite,,POINT(34.79643269660636 0.954060045767657),442,Non-crop +443,443,37.64560569646061,-0.4995748989339322,adadebay@umd.edu,false,2023-03-08 17:59,13.2 secs,Mapbox Satellite,,POINT(37.64560569646061 -0.499574898933932),443,Crop +444,444,35.193697555577735,-1.0090949072315327,aasareansah@gmail.com,false,2023-03-07 16:16,37.0 secs,Planet Monthly Mosaics,,POINT(35.193697555577735 -1.009094907231533),444,Non-crop +445,445,34.29444546033131,-0.1648308472018498,adadebay@umd.edu,false,2023-03-08 18:00,24.5 secs,Mapbox Satellite,,POINT(34.29444546033131 -0.16483084720185),445,Crop +446,446,34.314974037204585,-0.8050913043294229,isha9a@umd.edu,false,2023-03-08 21:03,61.4 secs,Mapbox Satellite,,POINT(34.314974037204585 -0.805091304329423),446,Non-crop +447,447,40.31294294731426,-1.7926326213420247,aasareansah@gmail.com,false,2023-03-07 16:17,13.0 secs,Planet Monthly Mosaics,,POINT(40.31294294731426 -1.792632621342025),447,Non-crop +448,448,36.91616303309522,-2.5727932209337974,aasareansah@gmail.com,false,2023-03-07 16:17,8.0 secs,Planet Monthly Mosaics,,POINT(36.91616303309522 -2.572793220933797),448,Non-crop +449,449,40.3485941388307,-1.8043211728138038,aasareansah@gmail.com,false,2023-03-07 16:18,63.2 secs,Planet Monthly Mosaics,,POINT(40.3485941388307 -1.804321172813804),449,Non-crop +450,450,34.75482976822808,1.2027323213729808,aasareansah@gmail.com,false,2023-03-07 16:18,24.0 secs,Planet Monthly Mosaics,,POINT(34.75482976822808 1.202732321372981),450,Crop +451,451,35.039639014761306,4.2483244000980145,aasareansah@gmail.com,false,2023-03-07 16:19,7.9 secs,Planet Monthly Mosaics,,POINT(35.039639014761306 4.248324400098014),451,Non-crop +452,452,40.2874788432159,2.504938312987063,aasareansah@gmail.com,false,2023-03-07 16:19,8.0 secs,Planet Monthly Mosaics,,POINT(40.2874788432159 2.504938312987063),452,Non-crop +453,453,37.573692021019895,-0.004673769817114266,aasareansah@gmail.com,false,2023-03-07 16:19,8.2 secs,Planet Monthly Mosaics,,POINT(37.573692021019895 -0.004673769817114),453,Non-crop +454,454,36.807037684141676,2.542819879774906,aasareansah@gmail.com,false,2023-03-07 16:19,10.9 secs,Planet Monthly Mosaics,,POINT(36.807037684141676 2.542819879774906),454,Non-crop +455,455,37.0048015872321,-0.45869277751116916,aasareansah@gmail.com,false,2023-03-07 16:19,16.3 secs,Planet Monthly Mosaics,,POINT(37.0048015872321 -0.458692777511169),455,Non-crop +456,456,36.84196187497421,3.7249805816569435,aasareansah@gmail.com,false,2023-03-07 16:19,8.3 secs,Planet Monthly Mosaics,,POINT(36.84196187497421 3.724980581656944),456,Non-crop +457,457,34.649646702806294,0.6912557299441862,isha9a@umd.edu,false,2023-03-08 21:07,203.2 secs,Mapbox Satellite,,POINT(34.649646702806294 0.691255729944186),457,Non-crop +458,458,39.27351047995854,2.382870821022328,aasareansah@gmail.com,false,2023-03-07 16:19,12.2 secs,Planet Monthly Mosaics,,POINT(39.27351047995854 2.382870821022328),458,Non-crop +459,459,34.09255046019023,0.13750458019816195,isha9a@umd.edu,false,2023-03-08 21:10,239.0 secs,Mapbox Satellite,,POINT(34.09255046019023 0.137504580198162),459,Non-crop +460,460,35.196367961673985,-0.9384964043103956,aasareansah@gmail.com,false,2023-03-07 16:20,12.1 secs,Planet Monthly Mosaics,,POINT(35.196367961673985 -0.938496404310396),460,Non-crop +461,461,34.729242476157864,-0.006280115060495967,isha9a@umd.edu,false,2023-03-08 21:16,331.4 secs,Planet Monthly Mosaics,,POINT(34.729242476157864 -0.006280115060496),461,Crop +462,462,37.14320247144663,-0.5799924116994661,aasareansah@gmail.com,false,2023-03-07 16:20,13.2 secs,Planet Monthly Mosaics,,POINT(37.14320247144663 -0.579992411699466),462,Non-crop +463,463,34.839386313724425,-0.5907161983375747,aasareansah@gmail.com,false,2023-03-07 16:20,14.5 secs,Planet Monthly Mosaics,,POINT(34.839386313724425 -0.590716198337575),463,Crop +464,464,37.298134606247274,-2.7530158736682773,aasareansah@gmail.com,false,2023-03-07 16:21,8.9 secs,Planet Monthly Mosaics,,POINT(37.298134606247274 -2.753015873668277),464,Non-crop +465,465,37.235080182010115,3.2939567840656823,aasareansah@gmail.com,false,2023-03-07 16:21,11.8 secs,Planet Monthly Mosaics,,POINT(37.235080182010115 3.293956784065682),465,Non-crop +466,466,34.620563644213945,3.1399987547652173,aasareansah@gmail.com,false,2023-03-07 16:21,9.6 secs,Planet Monthly Mosaics,,POINT(34.620563644213945 3.139998754765217),466,Non-crop +467,467,37.90330112445743,2.0453183232955094,aasareansah@gmail.com,false,2023-03-07 16:21,9.5 secs,Planet Monthly Mosaics,,POINT(37.90330112445743 2.045318323295509),467,Non-crop +468,468,38.75219866912606,1.4087696247600232,aasareansah@gmail.com,false,2023-03-07 16:21,8.3 secs,Planet Monthly Mosaics,,POINT(38.75219866912606 1.408769624760023),468,Non-crop +469,469,36.09478312311987,-0.45270767372427245,aasareansah@gmail.com,false,2023-03-07 16:21,13.3 secs,Planet Monthly Mosaics,,POINT(36.09478312311987 -0.452707673724272),469,Non-crop +470,470,39.14544637042974,3.0526918285416156,aasareansah@gmail.com,false,2023-03-07 16:21,9.1 secs,Planet Monthly Mosaics,,POINT(39.14544637042974 3.052691828541616),470,Non-crop +471,471,36.986050045933645,0.8166083567094135,aasareansah@gmail.com,false,2023-03-07 16:21,11.7 secs,Planet Monthly Mosaics,,POINT(36.986050045933645 0.816608356709414),471,Non-crop +472,472,36.67508229636412,2.360931840234958,aasareansah@gmail.com,false,2023-03-07 16:22,8.3 secs,Planet Monthly Mosaics,,POINT(36.67508229636412 2.360931840234958),472,Non-crop +473,473,34.87929322199816,-1.3410826721831222,aasareansah@gmail.com,false,2023-03-07 16:22,7.9 secs,Planet Monthly Mosaics,,POINT(34.87929322199816 -1.341082672183122),473,Non-crop +474,474,36.96034269804192,0.6752585022986302,aasareansah@gmail.com,false,2023-03-07 16:22,9.5 secs,Planet Monthly Mosaics,,POINT(36.96034269804192 0.67525850229863),474,Non-crop +475,475,38.01094738788046,0.4034110699662577,aasareansah@gmail.com,false,2023-03-07 16:22,11.6 secs,Planet Monthly Mosaics,,POINT(38.01094738788046 0.403411069966258),475,Non-crop +476,476,37.678195029831784,0.4669841625774332,aasareansah@gmail.com,false,2023-03-07 16:22,11.0 secs,Planet Monthly Mosaics,,POINT(37.678195029831784 0.466984162577433),476,Non-crop +477,477,35.56489986263098,4.17275389765291,aasareansah@gmail.com,false,2023-03-07 16:22,8.1 secs,Planet Monthly Mosaics,,POINT(35.56489986263098 4.17275389765291),477,Non-crop +478,478,38.89991542855711,-0.10599067675841689,aasareansah@gmail.com,false,2023-03-07 16:22,8.6 secs,Planet Monthly Mosaics,,POINT(38.89991542855711 -0.105990676758417),478,Non-crop +479,479,38.16088744929678,-0.5340005076082127,isha9a@umd.edu,false,2023-03-08 21:19,180.5 secs,Planet Monthly Mosaics,,POINT(38.16088744929678 -0.534000507608213),479,Crop +480,480,36.31334231145933,1.7564825870660954,aasareansah@gmail.com,false,2023-03-07 16:22,8.4 secs,Planet Monthly Mosaics,,POINT(36.31334231145933 1.756482587066095),480,Non-crop +481,481,36.05446021526722,-0.7701864509339067,isha9a@umd.edu,false,2023-03-08 21:24,327.0 secs,Planet Monthly Mosaics,,POINT(36.05446021526722 -0.770186450933907),481,Non-crop +482,482,34.634601312187485,0.08943773979142051,isha9a@umd.edu,false,2023-03-08 21:28,228.9 secs,Mapbox Satellite,,POINT(34.634601312187485 0.089437739791421),482,Non-crop +483,483,37.91044452150111,3.5311304324163824,aasareansah@gmail.com,false,2023-03-07 16:23,10.6 secs,Planet Monthly Mosaics,,POINT(37.91044452150111 3.531130432416382),483,Non-crop +484,484,34.5575298635729,-1.2831810660561516,isha9a@umd.edu,false,2023-03-08 21:29,71.7 secs,Mapbox Satellite,,POINT(34.5575298635729 -1.283181066056152),484,Crop +485,485,35.39516029245672,-0.8794258046257776,isha9a@umd.edu,false,2023-03-08 21:32,159.1 secs,Planet Monthly Mosaics,,POINT(35.39516029245672 -0.879425804625778),485,Non-crop +486,486,38.321928188337154,3.043383821526915,aasareansah@gmail.com,false,2023-03-07 16:24,14.3 secs,Planet Monthly Mosaics,,POINT(38.321928188337154 3.043383821526915),486,Non-crop +487,487,39.34805366474005,0.5760758805466408,aasareansah@gmail.com,false,2023-03-07 16:24,7.9 secs,Planet Monthly Mosaics,,POINT(39.34805366474005 0.576075880546641),487,Non-crop +488,488,40.70228392102145,2.2930049702457507,aasareansah@gmail.com,false,2023-03-07 16:24,11.6 secs,Planet Monthly Mosaics,,POINT(40.70228392102145 2.293004970245751),488,Non-crop +489,489,34.988652426404,2.685077579713715,aasareansah@gmail.com,false,2023-03-07 16:25,10.8 secs,Planet Monthly Mosaics,,POINT(34.988652426404 2.685077579713715),489,Non-crop +490,490,35.0036305876314,-0.7040441877978902,isha9a@umd.edu,false,2023-03-08 21:35,151.2 secs,Mapbox Satellite,,POINT(35.0036305876314 -0.70404418779789),490,Non-crop +491,491,34.653443139440675,-0.02515374865425375,aasareansah@gmail.com,false,2023-03-07 16:25,13.7 secs,Planet Monthly Mosaics,,POINT(34.653443139440675 -0.025153748654254),491,Non-crop +492,492,40.614414543790275,0.3435724730712614,aasareansah@gmail.com,false,2023-03-07 16:25,7.9 secs,Planet Monthly Mosaics,,POINT(40.614414543790275 0.343572473071261),492,Non-crop +493,493,38.51702418202515,-2.8238115957339627,aasareansah@gmail.com,false,2023-03-07 16:25,8.9 secs,Planet Monthly Mosaics,,POINT(38.51702418202515 -2.823811595733963),493,Non-crop +494,494,34.383934200549206,4.452384575378512,aasareansah@gmail.com,false,2023-03-07 16:25,8.8 secs,Planet Monthly Mosaics,,POINT(34.383934200549206 4.452384575378512),494,Non-crop +495,495,37.70530305115963,2.5868444101204076,aasareansah@gmail.com,false,2023-03-07 16:25,7.2 secs,Planet Monthly Mosaics,,POINT(37.70530305115963 2.586844410120408),495,Non-crop +496,496,35.33655489953117,-1.367002453719124,aasareansah@gmail.com,false,2023-03-07 16:25,7.7 secs,Planet Monthly Mosaics,,POINT(35.33655489953117 -1.367002453719124),496,Non-crop +497,497,38.39506049495438,2.5714102466385724,aasareansah@gmail.com,false,2023-03-07 16:25,10.1 secs,Planet Monthly Mosaics,,POINT(38.39506049495438 2.571410246638572),497,Non-crop +498,498,39.084849466631404,3.2272009282858454,aasareansah@gmail.com,false,2023-03-07 16:26,8.5 secs,Planet Monthly Mosaics,,POINT(39.084849466631404 3.227200928285845),498,Non-crop +499,499,37.09093189496214,0.9915204603295135,aasareansah@gmail.com,false,2023-03-07 16:26,8.0 secs,Planet Monthly Mosaics,,POINT(37.09093189496214 0.991520460329514),499,Non-crop +500,500,37.349874791030274,-2.6561425061602493,aasareansah@gmail.com,false,2023-03-07 16:26,13.6 secs,Planet Monthly Mosaics,,POINT(37.349874791030274 -2.656142506160249),500,Non-crop +501,501,35.17501834522299,-0.7515855198732806,isha9a@umd.edu,false,2023-03-08 21:36,80.3 secs,Sentinel-2,,POINT(35.17501834522299 -0.751585519873281),501,Non-crop +502,502,35.71291563450446,-0.6426003621868701,aasareansah@gmail.com,false,2023-03-07 16:26,10.2 secs,Planet Monthly Mosaics,,POINT(35.71291563450446 -0.64260036218687),502,Non-crop +503,503,34.73484872536813,-0.8482526044699542,isha9a@umd.edu,false,2023-03-08 21:38,103.7 secs,Mapbox Satellite,,POINT(34.73484872536813 -0.848252604469954),503,Crop +504,504,39.023752295479596,-1.6523111301971936,aasareansah@gmail.com,false,2023-03-07 16:27,8.9 secs,Planet Monthly Mosaics,,POINT(39.023752295479596 -1.652311130197194),504,Non-crop +505,505,40.78764234677859,2.693632786095179,aasareansah@gmail.com,false,2023-03-07 16:27,8.7 secs,Planet Monthly Mosaics,,POINT(40.78764234677859 2.693632786095179),505,Non-crop +506,506,35.06000147333601,-0.7966242423341563,aasareansah@gmail.com,false,2023-03-07 16:27,8.8 secs,Planet Monthly Mosaics,,POINT(35.06000147333601 -0.796624242334156),506,Non-crop +507,507,36.30009993341707,-0.8232024117545933,aasareansah@gmail.com,false,2023-03-07 16:27,12.8 secs,Planet Monthly Mosaics,,POINT(36.30009993341707 -0.823202411754593),507,Non-crop +508,508,35.10989219436885,-0.32787050503899345,aasareansah@gmail.com,false,2023-03-07 16:27,17.7 secs,Planet Monthly Mosaics,,POINT(35.10989219436885 -0.327870505038993),508,Crop +509,509,34.828144212308935,-0.5773129039263161,aasareansah@gmail.com,false,2023-03-07 16:27,11.8 secs,Planet Monthly Mosaics,,POINT(34.828144212308935 -0.577312903926316),509,Crop +510,510,35.97078220549668,2.9560655204795507,aasareansah@gmail.com,false,2023-03-07 16:27,10.1 secs,Planet Monthly Mosaics,,POINT(35.97078220549668 2.956065520479551),510,Non-crop +511,511,35.10284238847304,-0.4966636896018876,aasareansah@gmail.com,false,2023-03-07 16:28,8.4 secs,Planet Monthly Mosaics,,POINT(35.10284238847304 -0.496663689601888),511,Non-crop +512,512,40.132913213558695,-3.146251638192202,aasareansah@gmail.com,false,2023-03-07 16:28,19.5 secs,Planet Monthly Mosaics,,POINT(40.132913213558695 -3.146251638192202),512,Non-crop +513,513,34.22388893299226,0.605241533460933,isha9a@umd.edu,false,2023-03-08 21:38,29.8 secs,Planet Monthly Mosaics,,POINT(34.22388893299226 0.605241533460933),513,Crop +514,514,38.96670779650444,-0.4182302700090625,aasareansah@gmail.com,false,2023-03-07 16:28,9.9 secs,Planet Monthly Mosaics,,POINT(38.96670779650444 -0.418230270009062),514,Non-crop +515,515,40.32965437189055,1.3051310381454755,aasareansah@gmail.com,false,2023-03-07 16:28,7.9 secs,Planet Monthly Mosaics,,POINT(40.32965437189055 1.305131038145476),515,Non-crop +516,516,34.96478463720174,-0.5794803589111308,isha9a@umd.edu,false,2023-03-08 21:43,286.6 secs,Planet Monthly Mosaics,,POINT(34.96478463720174 -0.579480358911131),516,Non-crop +517,517,40.96226183600945,-0.2882607556531856,aasareansah@gmail.com,false,2023-03-07 16:29,12.1 secs,Planet Monthly Mosaics,,POINT(40.96226183600945 -0.288260755653186),517,Non-crop +518,518,39.735173373793074,3.116710754764235,aasareansah@gmail.com,false,2023-03-07 16:29,9.5 secs,Planet Monthly Mosaics,,POINT(39.735173373793074 3.116710754764235),518,Non-crop +519,519,36.82035819397363,1.9556658604114368,isha9a@umd.edu,false,2023-03-08 21:43,31.8 secs,Mapbox Satellite,,POINT(36.82035819397363 1.955665860411437),519,Non-crop +520,520,35.35383455962234,2.211003394302324,isha9a@umd.edu,false,2023-03-08 21:44,16.3 secs,Mapbox Satellite,,POINT(35.35383455962234 2.211003394302324),520,Non-crop +521,521,35.410735825610544,-1.428738022679518,isha9a@umd.edu,false,2023-03-08 21:44,12.7 secs,Mapbox Satellite,,POINT(35.410735825610544 -1.428738022679518),521,Non-crop +522,522,37.19939040219627,2.0807162473361056,isha9a@umd.edu,false,2023-03-08 21:44,31.6 secs,Mapbox Satellite,,POINT(37.19939040219627 2.080716247336106),522,Non-crop +523,523,35.08208831520865,-0.9479913632950039,isha9a@umd.edu,false,2023-03-08 21:45,48.4 secs,Planet Monthly Mosaics,,POINT(35.08208831520865 -0.947991363295004),523,Crop +524,524,36.529750333470275,-1.2510182266385572,isha9a@umd.edu,false,2023-03-08 21:46,33.4 secs,Mapbox Satellite,,POINT(36.529750333470275 -1.251018226638557),524,Non-crop +525,525,34.93545090273626,1.3595428403458083,isha9a@umd.edu,false,2023-03-08 21:46,43.9 secs,Planet Monthly Mosaics,,POINT(34.93545090273626 1.359542840345808),525,Non-crop +526,526,34.71477847358641,0.00774202647017627,isha9a@umd.edu,false,2023-03-08 21:48,86.0 secs,Mapbox Satellite,,POINT(34.71477847358641 0.007742026470176),526,Crop +527,527,39.59186188819468,-1.1566710064881998,isha9a@umd.edu,false,2023-03-08 21:49,49.0 secs,Mapbox Satellite,,POINT(39.59186188819468 -1.1566710064882),527,Non-crop +528,528,38.23861965981843,0.33277349267400996,isha9a@umd.edu,false,2023-03-08 21:49,18.7 secs,Mapbox Satellite,,POINT(38.23861965981843 0.33277349267401),528,Non-crop +529,529,37.01526073538915,-0.7368748435057784,isha9a@umd.edu,false,2023-03-08 21:51,95.2 secs,Mapbox Satellite,,POINT(37.01526073538915 -0.736874843505778),529,Non-crop +530,530,38.09145658119295,-2.415198363979528,isha9a@umd.edu,false,2023-03-08 21:56,336.6 secs,Planet Monthly Mosaics,,POINT(38.09145658119295 -2.415198363979528),530,Crop +531,531,37.65093779234663,-0.026120886979007708,isha9a@umd.edu,false,2023-03-08 21:57,61.5 secs,Mapbox Satellite,,POINT(37.65093779234663 -0.026120886979008),531,Non-crop +532,532,40.536095203749554,3.632313062717191,isha9a@umd.edu,false,2023-03-08 21:59,114.1 secs,Mapbox Satellite,,POINT(40.536095203749554 3.632313062717191),532,Non-crop +533,533,34.56842069799163,0.8483574983470237,isha9a@umd.edu,false,2023-03-08 22:04,293.7 secs,Mapbox Satellite,,POINT(34.56842069799163 0.848357498347024),533,Crop +534,534,34.560156122324,-0.01247945700671197,isha9a@umd.edu,false,2023-03-08 22:04,12.8 secs,Mapbox Satellite,,POINT(34.560156122324 -0.012479457006712),534,Crop +535,535,36.52662305326271,1.9856940678472765,isha9a@umd.edu,false,2023-03-08 22:05,31.0 secs,Planet Monthly Mosaics,,POINT(36.52662305326271 1.985694067847276),535,Non-crop +536,536,36.814465210482616,-1.0359101389648762,isha9a@umd.edu,false,2023-03-08 22:06,93.2 secs,Mapbox Satellite,,POINT(36.814465210482616 -1.035910138964876),536,Non-crop +537,537,37.06018956428111,-0.4699689113578657,isha9a@umd.edu,false,2023-03-08 22:07,22.5 secs,Mapbox Satellite,,POINT(37.06018956428111 -0.469968911357866),537,Non-crop +538,538,39.14612330417667,2.7377571101253593,isha9a@umd.edu,false,2023-03-08 22:07,28.3 secs,Planet Monthly Mosaics,,POINT(39.14612330417667 2.737757110125359),538,Non-crop +539,539,37.87082220460906,-1.7089443100333617,isha9a@umd.edu,false,2023-03-08 22:08,30.8 secs,Mapbox Satellite,,POINT(37.87082220460906 -1.708944310033362),539,Non-crop +540,540,34.88255928203005,-0.023363158733180508,isha9a@umd.edu,false,2023-03-08 22:08,23.5 secs,Mapbox Satellite,,POINT(34.88255928203005 -0.023363158733181),540,Non-crop +541,541,34.46410083114127,-0.8173018845840876,isha9a@umd.edu,false,2023-03-08 22:09,77.0 secs,Mapbox Satellite,,POINT(34.46410083114127 -0.817301884584088),541,Crop +542,542,34.93492776673225,2.8648969176507055,isha9a@umd.edu,false,2023-03-08 22:09,10.7 secs,Mapbox Satellite,,POINT(34.93492776673225 2.864896917650706),542,Non-crop +543,543,38.81338965106459,1.119704331352858,isha9a@umd.edu,false,2023-03-08 22:09,9.2 secs,Mapbox Satellite,,POINT(38.81338965106459 1.119704331352858),543,Non-crop \ No newline at end of file diff --git a/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv new file mode 100644 index 00000000..44018a52 --- /dev/null +++ b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv @@ -0,0 +1,545 @@ +plotid,sampleid,lon,lat,email,flagged,collection_time,analysis_duration,imagery_title,imagery_attributions,sample_geom,pl_sampleid,Does this point correspond to active cropland? +0,0,34.988629829555656,2.403552919494441,adadebay@umd.edu,false,2023-03-06 13:50,5.3 secs,,,POINT(34.988629829555656 2.403552919494441),0,Non-crop +1,1,40.4369134463712,0.23745156614928092,adadebay@umd.edu,false,2023-03-06 13:50,6.3 secs,,,POINT(40.4369134463712 0.237451566149281),1,Non-crop +2,2,35.525785266678824,4.203260690691489,adadebay@umd.edu,false,2023-03-06 13:50,21.8 secs,,,POINT(35.525785266678824 4.203260690691489),2,Non-crop +3,3,35.34502880181526,1.7861325423989396,adadebay@umd.edu,false,2023-02-28 20:29,8.1 secs,Mapbox Satellite,,POINT(35.34502880181526 1.78613254239894),3,Non-crop +4,4,41.092915614480624,-2.088706573827987,adadebay@umd.edu,false,2023-02-28 20:30,29.2 secs,Mapbox Satellite,,POINT(41.092915614480624 -2.088706573827987),4,Crop +5,5,34.4249103574835,-1.030559071942206,adadebay@umd.edu,false,2023-02-28 20:31,75.8 secs,Mapbox Satellite,,POINT(34.4249103574835 -1.030559071942206),5,Crop +6,6,39.74092423396898,0.4906842425498013,adadebay@umd.edu,false,2023-02-28 20:31,33.5 secs,Mapbox Satellite,,POINT(39.74092423396898 0.490684242549801),6,Non-crop +7,7,35.17727471050547,0.8244599235574998,adadebay@umd.edu,false,2023-02-28 20:32,26.9 secs,Planet Monthly Mosaics,,POINT(35.17727471050547 0.8244599235575),7,Crop +8,8,39.45538305881567,1.1210267423453937,adadebay@umd.edu,false,2023-02-28 20:32,8.4 secs,Planet Monthly Mosaics,,POINT(39.45538305881567 1.121026742345394),8,Non-crop +9,9,39.369209788785,-2.6825226398269617,adadebay@umd.edu,false,2023-02-28 20:32,13.6 secs,Planet Monthly Mosaics,,POINT(39.369209788785 -2.682522639826962),9,Non-crop +10,10,34.7476597030811,3.302863534933985,dianafrimpong710@gmail.com,false,2023-03-01 20:12,105.2 secs,Planet Monthly Mosaics,,POINT(34.7476597030811 3.302863534933985),10,Non-crop +11,11,37.52023980882631,1.1294265556977394,adadebay@umd.edu,false,2023-02-28 20:33,10.5 secs,Planet Monthly Mosaics,,POINT(37.52023980882631 1.129426555697739),11,Non-crop +12,12,40.04232638214461,0.10707860224488248,adadebay@umd.edu,false,2023-02-28 20:33,8.9 secs,Planet Monthly Mosaics,,POINT(40.04232638214461 0.107078602244882),12,Non-crop +13,13,36.59504709600916,2.079095336922312,adadebay@umd.edu,false,2023-02-28 20:33,7.9 secs,Planet Monthly Mosaics,,POINT(36.59504709600916 2.079095336922312),13,Non-crop +14,14,35.355886022749836,-1.7191142947031781,adadebay@umd.edu,false,2023-02-28 20:33,23.5 secs,Mapbox Satellite,,POINT(35.355886022749836 -1.719114294703178),14,Non-crop +15,15,37.94355071711686,0.2736254374888986,adadebay@umd.edu,false,2023-02-28 20:34,8.4 secs,Mapbox Satellite,,POINT(37.94355071711686 0.273625437488899),15,Crop +16,16,35.567931891106554,-0.963807053974099,adadebay@umd.edu,false,2023-02-28 20:38,278.5 secs,Planet Monthly Mosaics,,POINT(35.567931891106554 -0.963807053974099),16,Crop +17,17,35.2370463385883,0.23539226037375338,adadebay@umd.edu,false,2023-02-28 20:39,34.4 secs,Planet Monthly Mosaics,,POINT(35.2370463385883 0.235392260373753),17,Crop +18,18,34.56192216282922,-0.5423313934257865,adadebay@umd.edu,false,2023-02-28 20:44,328.6 secs,Planet Monthly Mosaics,,POINT(34.56192216282922 -0.542331393425786),18,Crop +19,19,39.91309533587102,0.5405029647849312,adadebay@umd.edu,false,2023-02-28 20:44,12.0 secs,Planet Monthly Mosaics,,POINT(39.91309533587102 0.540502964784931),19,Non-crop +20,20,37.77622013677315,-3.216077832453689,adadebay@umd.edu,false,2023-02-28 20:45,13.0 secs,Planet Monthly Mosaics,,POINT(37.77622013677315 -3.216077832453689),20,Non-crop +21,21,36.952890064644144,-0.7336035276099032,adadebay@umd.edu,false,2023-02-28 20:50,279.6 secs,Mapbox Satellite,,POINT(36.952890064644144 -0.733603527609903),21,Crop +22,22,40.81069538939396,3.8142341143013847,adadebay@umd.edu,false,2023-02-28 20:50,11.1 secs,Mapbox Satellite,,POINT(40.81069538939396 3.814234114301385),22,Non-crop +23,23,39.31291565564592,-4.292596583003278,adadebay@umd.edu,false,2023-02-28 20:51,19.5 secs,Mapbox Satellite,,POINT(39.31291565564592 -4.292596583003278),23,Non-crop +24,24,36.56392047775658,-1.4401680504539716,adadebay@umd.edu,false,2023-02-28 20:51,12.1 secs,Mapbox Satellite,,POINT(36.56392047775658 -1.440168050453972),24,Non-crop +25,25,40.761482926934235,1.797142812318858,adadebay@umd.edu,false,2023-02-28 20:51,7.1 secs,Mapbox Satellite,,POINT(40.761482926934235 1.797142812318858),25,Non-crop +26,26,35.157061444706144,1.0119532740001498,adadebay@umd.edu,false,2023-02-28 20:52,61.0 secs,Planet Monthly Mosaics,,POINT(35.157061444706144 1.01195327400015),26,Non-crop +27,27,34.203510614976835,3.843240810122485,adadebay@umd.edu,false,2023-02-28 20:52,11.2 secs,Planet Monthly Mosaics,,POINT(34.203510614976835 3.843240810122485),27,Non-crop +28,28,40.634491221723714,1.4292896941865394,adadebay@umd.edu,false,2023-02-28 20:52,8.5 secs,Planet Monthly Mosaics,,POINT(40.634491221723714 1.429289694186539),28,Non-crop +29,29,37.24315858180454,0.18885273302287983,adadebay@umd.edu,false,2023-02-28 20:53,19.8 secs,Mapbox Satellite,,POINT(37.24315858180454 0.18885273302288),29,Non-crop +30,30,38.50529869815832,1.3266174643205182,adadebay@umd.edu,false,2023-02-28 20:53,9.2 secs,Mapbox Satellite,,POINT(38.50529869815832 1.326617464320518),30,Non-crop +31,31,40.02570086581538,2.052485246929315,adadebay@umd.edu,false,2023-02-28 20:53,17.0 secs,Mapbox Satellite,,POINT(40.02570086581538 2.052485246929315),31,Non-crop +32,32,36.08923467501349,2.8152333378024177,adadebay@umd.edu,false,2023-02-28 20:53,8.1 secs,Mapbox Satellite,,POINT(36.08923467501349 2.815233337802418),32,Non-crop +33,33,38.25468798085764,-3.293961403681881,adadebay@umd.edu,false,2023-02-28 20:53,7.1 secs,Mapbox Satellite,,POINT(38.25468798085764 -3.293961403681881),33,Non-crop +34,34,36.24856761389988,2.978220395675694,adadebay@umd.edu,false,2023-02-28 20:53,6.4 secs,Mapbox Satellite,,POINT(36.24856761389988 2.978220395675694),34,Non-crop +35,35,35.08856440679308,-0.610412869850806,adadebay@umd.edu,false,2023-02-28 20:53,9.1 secs,Mapbox Satellite,,POINT(35.08856440679308 -0.610412869850806),35,Non-crop +36,36,37.10586175136883,0.6450479726392292,adadebay@umd.edu,false,2023-02-28 20:54,14.6 secs,Mapbox Satellite,,POINT(37.10586175136883 0.645047972639229),36,Non-crop +37,37,36.28157017849977,-0.844978485851314,adadebay@umd.edu,false,2023-02-28 20:54,12.6 secs,Mapbox Satellite,,POINT(36.28157017849977 -0.844978485851314),37,Crop +38,38,35.24641350935284,3.1088112936312293,adadebay@umd.edu,false,2023-02-28 20:54,12.1 secs,Mapbox Satellite,,POINT(35.24641350935284 3.108811293631229),38,Non-crop +39,39,35.725864381252485,-0.9901508750054767,adadebay@umd.edu,false,2023-02-28 20:54,7.7 secs,Mapbox Satellite,,POINT(35.725864381252485 -0.990150875005477),39,Crop +40,40,36.68065734363421,-0.8149776658973387,adadebay@umd.edu,false,2023-02-28 20:54,7.6 secs,Mapbox Satellite,,POINT(36.68065734363421 -0.814977665897339),40,Non-crop +41,41,39.78768478044772,3.532698290163748,adadebay@umd.edu,false,2023-02-28 20:54,7.8 secs,Mapbox Satellite,,POINT(39.78768478044772 3.532698290163748),41,Non-crop +42,42,34.937338523086936,1.1808045765140807,adadebay@umd.edu,false,2023-02-28 20:55,8.7 secs,Mapbox Satellite,,POINT(34.937338523086936 1.180804576514081),42,Crop +43,43,35.95117962367605,0.11765502410130428,adadebay@umd.edu,false,2023-02-28 20:55,37.5 secs,Planet Monthly Mosaics,,POINT(35.95117962367605 0.117655024101304),43,Crop +44,44,36.233816518951166,-0.9529835747040568,adadebay@umd.edu,false,2023-02-28 20:56,32.4 secs,Mapbox Satellite,,POINT(36.233816518951166 -0.952983574704057),44,Non-crop +45,45,40.34542919363465,-0.16642752812001055,adadebay@umd.edu,false,2023-02-28 20:56,6.8 secs,Mapbox Satellite,,POINT(40.34542919363465 -0.166427528120011),45,Non-crop +46,46,35.109574883126434,1.7949982414224284,adadebay@umd.edu,false,2023-02-28 20:56,7.2 secs,Mapbox Satellite,,POINT(35.109574883126434 1.794998241422428),46,Non-crop +47,47,38.10104270259982,2.2770658661008274,adadebay@umd.edu,false,2023-02-28 20:56,7.5 secs,Mapbox Satellite,,POINT(38.10104270259982 2.277065866100827),47,Non-crop +48,48,34.93916270476336,-0.5632779940672789,adadebay@umd.edu,false,2023-02-28 20:56,21.3 secs,Mapbox Satellite,,POINT(34.93916270476336 -0.563277994067279),48,Non-crop +49,49,35.698150869472485,2.656619125017031,adadebay@umd.edu,false,2023-02-28 20:57,9.1 secs,Mapbox Satellite,,POINT(35.698150869472485 2.656619125017031),49,Non-crop +50,50,38.332830459210996,-2.4627477963661004,adadebay@umd.edu,false,2023-02-28 20:57,8.2 secs,Mapbox Satellite,,POINT(38.332830459210996 -2.4627477963661),50,Non-crop +51,51,36.384083118334274,4.384630880395623,adadebay@umd.edu,false,2023-02-28 20:57,8.3 secs,Mapbox Satellite,,POINT(36.384083118334274 4.384630880395623),51,Non-crop +52,52,35.01471998180397,-0.42170819623595546,adadebay@umd.edu,false,2023-02-28 20:57,19.5 secs,Mapbox Satellite,,POINT(35.01471998180397 -0.421708196235955),52,Non-crop +53,53,37.47142387421962,2.60955944432736,adadebay@umd.edu,false,2023-02-28 20:57,8.9 secs,Mapbox Satellite,,POINT(37.47142387421962 2.60955944432736),53,Non-crop +54,54,37.44533838762773,-1.8552448087267317,adadebay@umd.edu,false,2023-02-28 20:58,15.7 secs,Mapbox Satellite,,POINT(37.44533838762773 -1.855244808726732),54,Non-crop +55,55,40.14514392119623,-3.098656053349216,adadebay@umd.edu,false,2023-02-28 21:00,149.6 secs,Mapbox Satellite,,POINT(40.14514392119623 -3.098656053349216),55,Non-crop +56,56,38.6594746226374,-3.835613952172189,adadebay@umd.edu,false,2023-02-28 21:00,11.1 secs,Mapbox Satellite,,POINT(38.6594746226374 -3.835613952172189),56,Non-crop +57,57,39.02562274024034,0.07077064319861262,adadebay@umd.edu,false,2023-02-28 21:00,7.3 secs,Mapbox Satellite,,POINT(39.02562274024034 0.070770643198613),57,Non-crop +58,58,36.02109758574353,3.5366108060174963,adadebay@umd.edu,false,2023-02-28 21:00,5.2 secs,Mapbox Satellite,,POINT(36.02109758574353 3.536610806017496),58,Non-crop +59,59,35.06865100448405,1.187649341007699,adadebay@umd.edu,false,2023-02-28 21:04,231.0 secs,Mapbox Satellite,,POINT(35.06865100448405 1.187649341007699),59,Non-crop +60,60,38.263410192690564,2.051383658830262,adadebay@umd.edu,false,2023-02-28 21:04,11.8 secs,Mapbox Satellite,,POINT(38.263410192690564 2.051383658830262),60,Non-crop +61,61,40.14898007195484,3.3226177399453816,adadebay@umd.edu,false,2023-02-28 21:05,16.1 secs,Mapbox Satellite,,POINT(40.14898007195484 3.322617739945382),61,Non-crop +62,62,37.206652528458754,-1.1530997062886132,adadebay@umd.edu,false,2023-02-28 21:05,8.7 secs,Mapbox Satellite,,POINT(37.206652528458754 -1.153099706288613),62,Crop +63,63,36.33551213412802,-2.0059902859724703,adadebay@umd.edu,false,2023-02-28 21:05,12.3 secs,Mapbox Satellite,,POINT(36.33551213412802 -2.00599028597247),63,Non-crop +64,64,37.81898292849584,2.890470600118884,adadebay@umd.edu,false,2023-02-28 21:05,11.1 secs,Mapbox Satellite,,POINT(37.81898292849584 2.890470600118884),64,Non-crop +65,65,36.964503175730776,-1.8728601229685733,adadebay@umd.edu,false,2023-02-28 21:06,17.4 secs,Mapbox Satellite,,POINT(36.964503175730776 -1.872860122968573),65,Non-crop +66,66,35.432462250513026,-0.21706777965244287,adadebay@umd.edu,false,2023-02-28 21:06,25.0 secs,Mapbox Satellite,,POINT(35.432462250513026 -0.217067779652443),66,Non-crop +67,67,35.12381890958649,0.7606469113885553,adadebay@umd.edu,false,2023-02-28 21:06,9.4 secs,Mapbox Satellite,,POINT(35.12381890958649 0.760646911388555),67,Non-crop +68,68,34.30929368936593,0.47748228001945475,adadebay@umd.edu,false,2023-02-28 21:06,13.4 secs,Mapbox Satellite,,POINT(34.30929368936593 0.477482280019455),68,Non-crop +69,69,36.6457343594037,-0.49640885252982253,adadebay@umd.edu,false,2023-02-28 21:06,11.1 secs,Mapbox Satellite,,POINT(36.6457343594037 -0.496408852529823),69,Non-crop +70,70,34.226388257459824,-0.6041420114577926,adadebay@umd.edu,false,2023-02-28 21:07,24.5 secs,Mapbox Satellite,,POINT(34.226388257459824 -0.604142011457793),70,Non-crop +71,71,38.249558852614086,3.1963730568049713,adadebay@umd.edu,false,2023-02-28 21:07,18.0 secs,Mapbox Satellite,,POINT(38.249558852614086 3.196373056804971),71,Non-crop +72,72,35.416853560011795,3.8915106166058218,adadebay@umd.edu,false,2023-02-28 21:07,12.3 secs,Mapbox Satellite,,POINT(35.416853560011795 3.891510616605822),72,Non-crop +73,73,34.793844712321096,1.0897067468642034,adadebay@umd.edu,false,2023-02-28 21:08,13.6 secs,Mapbox Satellite,,POINT(34.793844712321096 1.089706746864203),73,Crop +74,74,35.82894420505845,0.4963590966445787,adadebay@umd.edu,false,2023-02-28 21:08,23.3 secs,Mapbox Satellite,,POINT(35.82894420505845 0.496359096644579),74,Non-crop +75,75,40.86680566262454,2.5510833756481532,adadebay@umd.edu,false,2023-02-28 21:08,21.3 secs,Mapbox Satellite,,POINT(40.86680566262454 2.551083375648153),75,Non-crop +76,76,37.42434808348648,2.172412965946544,adadebay@umd.edu,false,2023-02-28 21:08,9.9 secs,Mapbox Satellite,,POINT(37.42434808348648 2.172412965946544),76,Non-crop +77,77,37.15129430727518,-0.7226093010817065,adadebay@umd.edu,false,2023-02-28 21:09,41.4 secs,Mapbox Satellite,,POINT(37.15129430727518 -0.722609301081706),77,Non-crop +78,78,35.24400762282487,-0.8344837781524341,adadebay@umd.edu,false,2023-02-28 21:10,38.8 secs,Mapbox Satellite,,POINT(35.24400762282487 -0.834483778152434),78,Non-crop +79,79,35.89383258961963,-0.3086155983480622,adadebay@umd.edu,false,2023-02-28 21:10,26.0 secs,Mapbox Satellite,,POINT(35.89383258961963 -0.308615598348062),79,Crop +80,80,34.8374317772852,1.2105020095980699,adadebay@umd.edu,false,2023-02-28 21:10,11.8 secs,Mapbox Satellite,,POINT(34.8374317772852 1.21050200959807),80,Crop +81,81,37.47851943218349,-1.5673861189284093,adadebay@umd.edu,false,2023-02-28 21:12,73.3 secs,Planet Monthly Mosaics,,POINT(37.47851943218349 -1.567386118928409),81,Crop +82,82,37.62528047925142,-3.386462077890213,adadebay@umd.edu,false,2023-02-28 21:12,19.3 secs,Mapbox Satellite,,POINT(37.62528047925142 -3.386462077890213),82,Non-crop +83,83,37.94051901411475,0.003977377954060247,adadebay@umd.edu,false,2023-02-28 21:13,32.1 secs,Mapbox Satellite,,POINT(37.94051901411475 0.00397737795406),83,Non-crop +84,84,38.7070591354366,-2.0059845289306772,adadebay@umd.edu,false,2023-02-28 21:13,8.9 secs,Mapbox Satellite,,POINT(38.7070591354366 -2.005984528930677),84,Non-crop +85,85,34.838779927856656,-0.86257077504155,adadebay@umd.edu,false,2023-02-28 21:13,10.0 secs,Mapbox Satellite,,POINT(34.838779927856656 -0.86257077504155),85,Crop +86,86,37.44660412263658,1.2193124117332885,adadebay@umd.edu,false,2023-02-28 21:13,7.3 secs,Mapbox Satellite,,POINT(37.44660412263658 1.219312411733288),86,Non-crop +87,87,34.6632601985172,0.07559985882881662,adadebay@umd.edu,false,2023-02-28 21:13,9.7 secs,Mapbox Satellite,,POINT(34.6632601985172 0.075599858828817),87,Non-crop +88,88,34.638949502003825,0.6729112014973999,adadebay@umd.edu,false,2023-02-28 21:13,8.3 secs,Mapbox Satellite,,POINT(34.638949502003825 0.6729112014974),88,Crop +89,89,34.5994023707549,0.7630158568229832,adadebay@umd.edu,false,2023-02-28 21:14,31.3 secs,Mapbox Satellite,,POINT(34.5994023707549 0.763015856822983),89,Crop +90,90,36.93706274251377,2.255505688906279,adadebay@umd.edu,false,2023-02-28 21:14,8.4 secs,Mapbox Satellite,,POINT(36.93706274251377 2.255505688906279),90,Non-crop +91,91,40.44223583641046,4.03082660667592,adadebay@umd.edu,false,2023-02-28 21:14,6.8 secs,Mapbox Satellite,,POINT(40.44223583641046 4.03082660667592),91,Non-crop +92,92,36.06039300271393,-0.5684581218619031,adadebay@umd.edu,false,2023-02-28 21:14,6.7 secs,Mapbox Satellite,,POINT(36.06039300271393 -0.568458121861903),92,Crop +93,93,37.59588686792248,0.11931919114708699,adadebay@umd.edu,false,2023-02-28 21:14,8.2 secs,Mapbox Satellite,,POINT(37.59588686792248 0.119319191147087),93,Non-crop +94,94,34.95006733497294,-0.7711616777128759,adadebay@umd.edu,false,2023-02-28 21:14,7.7 secs,Mapbox Satellite,,POINT(34.95006733497294 -0.771161677712876),94,Crop +95,95,34.37280204804135,-0.5213450533332684,adadebay@umd.edu,false,2023-02-28 21:15,11.7 secs,Mapbox Satellite,,POINT(34.37280204804135 -0.521345053333268),95,Non-crop +96,96,35.92793233525755,-2.0132818618461816,adadebay@umd.edu,false,2023-02-28 21:15,8.8 secs,Mapbox Satellite,,POINT(35.92793233525755 -2.013281861846182),96,Non-crop +97,97,40.50489472621563,-1.7350862750790679,adadebay@umd.edu,false,2023-02-28 21:15,9.1 secs,Mapbox Satellite,,POINT(40.50489472621563 -1.735086275079068),97,Non-crop +98,98,34.84343325317191,-0.5513431781255153,adadebay@umd.edu,false,2023-02-28 21:15,9.0 secs,Mapbox Satellite,,POINT(34.84343325317191 -0.551343178125515),98,Crop +99,99,38.810924566797794,2.4564363287855104,adadebay@umd.edu,false,2023-02-28 21:15,6.4 secs,Mapbox Satellite,,POINT(38.810924566797794 2.45643632878551),99,Non-crop +100,100,38.82801401094352,0.44583460966597566,adadebay@umd.edu,false,2023-02-28 21:15,5.4 secs,Mapbox Satellite,,POINT(38.82801401094352 0.445834609665976),100,Non-crop +101,101,35.01842360551604,1.951978852908563,adadebay@umd.edu,false,2023-02-28 21:15,5.8 secs,Mapbox Satellite,,POINT(35.01842360551604 1.951978852908563),101,Non-crop +102,102,38.63846258841169,-1.8744628832696686,adadebay@umd.edu,false,2023-02-28 21:15,8.7 secs,Mapbox Satellite,,POINT(38.63846258841169 -1.874462883269669),102,Non-crop +103,103,34.67637175915808,4.509081535107014,adadebay@umd.edu,false,2023-02-28 21:16,10.6 secs,Mapbox Satellite,,POINT(34.67637175915808 4.509081535107014),103,Non-crop +104,104,34.80059714936316,-0.6330272098164963,adadebay@umd.edu,false,2023-02-28 21:16,40.0 secs,Mapbox Satellite,,POINT(34.80059714936316 -0.633027209816496),104,Crop +105,105,39.03989200445923,-2.346187888919231,adadebay@umd.edu,false,2023-02-28 21:16,14.4 secs,Mapbox Satellite,,POINT(39.03989200445923 -2.346187888919231),105,Non-crop +106,106,37.75101158485352,-0.05595061740159739,adadebay@umd.edu,false,2023-02-28 21:17,24.9 secs,Mapbox Satellite,,POINT(37.75101158485352 -0.055950617401597),106,Non-crop +107,107,34.94705324098189,-0.5166194835899087,adadebay@umd.edu,false,2023-02-28 21:19,141.7 secs,Mapbox Satellite,,POINT(34.94705324098189 -0.516619483589909),107,Non-crop +108,108,36.43853260211107,3.97061768786652,adadebay@umd.edu,false,2023-02-28 21:19,14.7 secs,Mapbox Satellite,,POINT(36.43853260211107 3.97061768786652),108,Non-crop +109,109,38.8861466691318,3.0445834421692615,adadebay@umd.edu,false,2023-02-28 21:20,8.8 secs,Mapbox Satellite,,POINT(38.8861466691318 3.044583442169262),109,Non-crop +110,110,35.876614895369954,3.7678862374880135,adadebay@umd.edu,false,2023-02-28 21:20,6.0 secs,Mapbox Satellite,,POINT(35.876614895369954 3.767886237488014),110,Non-crop +111,111,38.82883825639645,2.5306209549193475,adadebay@umd.edu,false,2023-02-28 21:20,7.2 secs,Mapbox Satellite,,POINT(38.82883825639645 2.530620954919348),111,Non-crop +112,112,38.2367270952654,-1.3301300190469625,adadebay@umd.edu,false,2023-02-28 21:20,8.6 secs,Mapbox Satellite,,POINT(38.2367270952654 -1.330130019046962),112,Non-crop +113,113,37.63554202932443,0.03299322115736923,adadebay@umd.edu,false,2023-02-28 21:21,88.1 secs,Mapbox Satellite,,POINT(37.63554202932443 0.032993221157369),113,Crop +114,114,37.05135107858711,-0.388438674158326,adadebay@umd.edu,false,2023-02-28 21:23,99.0 secs,Mapbox Satellite,,POINT(37.05135107858711 -0.388438674158326),114,Crop +115,115,38.39683562140306,-3.562375659885144,adadebay@umd.edu,false,2023-02-28 21:23,9.3 secs,Mapbox Satellite,,POINT(38.39683562140306 -3.562375659885144),115,Crop +116,116,38.47807537401632,-1.3194672935579193,adadebay@umd.edu,false,2023-02-28 21:23,8.1 secs,Mapbox Satellite,,POINT(38.47807537401632 -1.319467293557919),116,Non-crop +117,117,37.90724334671853,1.2126009407086038,adadebay@umd.edu,false,2023-02-28 21:24,13.1 secs,Mapbox Satellite,,POINT(37.90724334671853 1.212600940708604),117,Non-crop +118,118,35.1493469381087,0.7857613956282058,adadebay@umd.edu,false,2023-02-28 21:24,11.7 secs,Mapbox Satellite,,POINT(35.1493469381087 0.785761395628206),118,Crop +119,119,37.04107541888969,2.415639616416417,adadebay@umd.edu,false,2023-02-28 21:24,9.1 secs,Mapbox Satellite,,POINT(37.04107541888969 2.415639616416417),119,Non-crop +120,120,37.135914893319935,0.8892938871269861,adadebay@umd.edu,false,2023-02-28 21:24,7.7 secs,Mapbox Satellite,,POINT(37.135914893319935 0.889293887126986),120,Non-crop +121,121,34.280251153267145,-0.8857188443090579,adadebay@umd.edu,false,2023-02-28 21:24,8.4 secs,Mapbox Satellite,,POINT(34.280251153267145 -0.885718844309058),121,Crop +122,122,37.104919656283776,0.944772577359767,adadebay@umd.edu,false,2023-02-28 21:24,9.7 secs,Mapbox Satellite,,POINT(37.104919656283776 0.944772577359767),122,Crop +123,123,36.91642788302333,-1.8734338149965883,adadebay@umd.edu,false,2023-02-28 21:24,9.6 secs,Mapbox Satellite,,POINT(36.91642788302333 -1.873433814996588),123,Non-crop +124,124,35.6530472090822,1.3337481691783033,adadebay@umd.edu,false,2023-02-28 21:25,10.5 secs,Mapbox Satellite,,POINT(35.6530472090822 1.333748169178303),124,Non-crop +125,125,37.665284200034286,-0.8743618235367431,adadebay@umd.edu,false,2023-02-28 21:34,590.8 secs,Mapbox Satellite,,POINT(37.665284200034286 -0.874361823536743),125,Non-crop +126,126,40.28931527278772,-1.6826194889901331,adadebay@umd.edu,false,2023-02-28 21:35,7.8 secs,Mapbox Satellite,,POINT(40.28931527278772 -1.682619488990133),126,Non-crop +127,127,35.24886337281288,0.9780628088660805,adadebay@umd.edu,false,2023-02-28 21:36,112.4 secs,Mapbox Satellite,,POINT(35.24886337281288 0.97806280886608),127,Crop +128,128,35.350804743305915,-0.14292567584629862,adadebay@umd.edu,false,2023-02-28 21:37,8.9 secs,Mapbox Satellite,,POINT(35.350804743305915 -0.142925675846299),128,Crop +129,129,36.875372636897794,-1.9178813516888036,adadebay@umd.edu,false,2023-02-28 21:37,13.2 secs,Mapbox Satellite,,POINT(36.875372636897794 -1.917881351688804),129,Non-crop +130,130,36.36696602805077,-1.0011713498446717,adadebay@umd.edu,false,2023-02-28 21:37,14.2 secs,Mapbox Satellite,,POINT(36.36696602805077 -1.001171349844672),130,Non-crop +131,131,34.03567997149679,-0.45059458138419733,adadebay@umd.edu,false,2023-02-28 21:37,10.0 secs,Mapbox Satellite,,POINT(34.03567997149679 -0.450594581384197),131,Non-crop +132,132,34.31052408060243,0.06011131997460628,adadebay@umd.edu,false,2023-02-28 21:37,16.5 secs,Mapbox Satellite,,POINT(34.31052408060243 0.060111319974606),132,Non-crop +133,133,40.112200611852735,3.057621068313114,adadebay@umd.edu,false,2023-03-01 20:11,11.3 secs,Mapbox Satellite,,POINT(40.112200611852735 3.057621068313114),133,Non-crop +134,134,37.59446156272681,-0.6348207832764996,adadebay@umd.edu,false,2023-03-01 20:12,39.1 secs,Mapbox Satellite,,POINT(37.59446156272681 -0.6348207832765),134,Non-crop +135,135,40.9961063143978,-1.6779669321206159,dianafrimpong710@gmail.com,false,2023-03-01 20:12,46.4 secs,Planet Monthly Mosaics,,POINT(40.9961063143978 -1.677966932120616),135,Non-crop +136,136,39.22212647015637,2.6922268323855563,adadebay@umd.edu,false,2023-03-01 20:12,5.6 secs,Mapbox Satellite,,POINT(39.22212647015637 2.692226832385556),136,Non-crop +137,137,36.21832154143681,-0.33166413329927724,adadebay@umd.edu,false,2023-03-01 20:12,49.2 secs,Planet Monthly Mosaics,,POINT(36.21832154143681 -0.331664133299277),137,Crop +138,138,34.51475222889342,-0.7822238210290815,adadebay@umd.edu,false,2023-03-01 20:13,35.1 secs,Mapbox Satellite,,POINT(34.51475222889342 -0.782223821029082),138,Crop +139,139,38.50386682037691,-0.7440523952127349,dianafrimpong710@gmail.com,false,2023-03-01 20:14,126.1 secs,Sentinel-2,,POINT(38.50386682037691 -0.744052395212735),139,Crop +140,140,34.544980328359806,3.3127536927240926,adadebay@umd.edu,false,2023-03-01 20:54,9.0 secs,Mapbox Satellite,,POINT(34.544980328359806 3.312753692724093),140,Non-crop +141,141,35.072614057925755,2.0515640292935595,dianafrimpong710@gmail.com,false,2023-03-01 20:15,40.5 secs,Planet Monthly Mosaics,,POINT(35.072614057925755 2.05156402929356),141,Non-crop +142,142,37.63731851488094,-0.7082741543121748,dianafrimpong710@gmail.com,false,2023-03-01 20:16,59.9 secs,Planet Monthly Mosaics,,POINT(37.63731851488094 -0.708274154312175),142,Non-crop +143,143,36.79356032174582,-1.3320484579772995,dianafrimpong710@gmail.com,false,2023-03-01 20:16,26.8 secs,Planet Monthly Mosaics,,POINT(36.79356032174582 -1.3320484579773),143,Non-crop +144,144,37.196987405380554,-0.997502551481028,adadebay@umd.edu,false,2023-03-01 20:54,10.5 secs,Mapbox Satellite,,POINT(37.196987405380554 -0.997502551481028),144,Crop +145,145,35.97558762159962,-0.12276447907124412,dianafrimpong710@gmail.com,false,2023-03-01 20:18,21.7 secs,Planet Monthly Mosaics,,POINT(35.97558762159962 -0.122764479071244),145,Crop +146,146,34.1891721541313,-0.6146176829171154,dianafrimpong710@gmail.com,false,2023-03-01 20:20,114.5 secs,Planet Monthly Mosaics,,POINT(34.1891721541313 -0.614617682917115),146,Non-crop +147,147,36.92847527942525,1.2782331263275961,dianafrimpong710@gmail.com,false,2023-03-01 20:21,18.2 secs,Planet Monthly Mosaics,,POINT(36.92847527942525 1.278233126327596),147,Non-crop +148,148,37.67795253138457,-0.29646401880195755,adadebay@umd.edu,false,2023-03-01 20:55,55.8 secs,Planet Monthly Mosaics,,POINT(37.67795253138457 -0.296464018801958),148,Non-crop +149,149,38.06462053176975,-1.043559942579206,dianafrimpong710@gmail.com,false,2023-03-01 20:24,110.5 secs,Planet Monthly Mosaics,,POINT(38.06462053176975 -1.043559942579206),149,Non-crop +150,150,34.886305931598756,1.5162521774427242,dianafrimpong710@gmail.com,false,2023-03-01 20:24,19.2 secs,Planet Monthly Mosaics,,POINT(34.886305931598756 1.516252177442724),150,Non-crop +151,151,36.96408686642167,0.4599662921332445,dianafrimpong710@gmail.com,false,2023-03-01 20:24,33.6 secs,Planet Monthly Mosaics,,POINT(36.96408686642167 0.459966292133244),151,Non-crop +152,152,35.13520286056352,-1.1297140056307589,dianafrimpong710@gmail.com,false,2023-03-01 20:26,101.1 secs,Sentinel-2,,POINT(35.13520286056352 -1.129714005630759),152,Crop +153,153,41.800681225973825,3.9420204760538824,dianafrimpong710@gmail.com,false,2023-03-01 20:26,30.5 secs,Planet Monthly Mosaics,,POINT(41.800681225973825 3.942020476053882),153,Non-crop +154,154,36.41868031095638,-1.6763774091068298,dianafrimpong710@gmail.com,false,2023-03-01 20:27,52.5 secs,Planet Monthly Mosaics,,POINT(36.41868031095638 -1.67637740910683),154,Non-crop +155,155,38.033900036116925,-1.3282559996083236,dianafrimpong710@gmail.com,false,2023-03-01 20:28,52.6 secs,Planet Monthly Mosaics,,POINT(38.033900036116925 -1.328255999608324),155,Non-crop +156,156,39.29015146450414,-4.3996825162698165,adadebay@umd.edu,false,2023-03-01 20:59,257.8 secs,Mapbox Satellite,,POINT(39.29015146450414 -4.399682516269816),156,Crop +157,157,37.44623545167376,-1.0728680549928664,dianafrimpong710@gmail.com,false,2023-03-01 20:32,66.2 secs,Planet Monthly Mosaics,,POINT(37.44623545167376 -1.072868054992866),157,Non-crop +158,158,40.00198076172898,-1.4034050963748328,dianafrimpong710@gmail.com,false,2023-03-01 20:34,128.8 secs,Sentinel-2,,POINT(40.00198076172898 -1.403405096374833),158,Crop +159,159,34.51540863900097,0.20085040451937342,dianafrimpong710@gmail.com,false,2023-03-01 20:36,133.9 secs,Sentinel-2,,POINT(34.51540863900097 0.200850404519373),159,Crop +160,160,34.364068159964674,-0.027030376910205087,dianafrimpong710@gmail.com,false,2023-03-01 20:37,73.3 secs,Sentinel-2,,POINT(34.364068159964674 -0.027030376910205),160,Crop +161,161,38.97895604283259,2.4386926635852486,dianafrimpong710@gmail.com,false,2023-03-01 20:38,26.4 secs,Planet Monthly Mosaics,,POINT(38.97895604283259 2.438692663585249),161,Non-crop +162,162,35.27387999609261,0.11199548140903959,dianafrimpong710@gmail.com,false,2023-03-01 20:38,19.0 secs,Planet Monthly Mosaics,,POINT(35.27387999609261 0.11199548140904),162,Crop +163,163,39.911542341177416,-2.5611092030692055,dianafrimpong710@gmail.com,false,2023-03-01 20:38,16.1 secs,Planet Monthly Mosaics,,POINT(39.911542341177416 -2.561109203069206),163,Non-crop +164,164,35.185079473396854,0.5617606376305225,dianafrimpong710@gmail.com,false,2023-03-01 20:39,44.7 secs,Planet Monthly Mosaics,,POINT(35.185079473396854 0.561760637630522),164,Crop +165,165,35.05688885441114,-0.6795407745893438,adadebay@umd.edu,false,2023-03-01 20:59,7.0 secs,Mapbox Satellite,,POINT(35.05688885441114 -0.679540774589344),165,Crop +166,166,37.58793262053442,3.7690288499729587,dianafrimpong710@gmail.com,false,2023-03-01 20:40,20.6 secs,Planet Monthly Mosaics,,POINT(37.58793262053442 3.769028849972959),166,Non-crop +167,167,34.45875563125551,0.6272795031255634,dianafrimpong710@gmail.com,false,2023-03-01 20:41,64.2 secs,Sentinel-2,,POINT(34.45875563125551 0.627279503125563),167,Crop +168,168,34.40926954593081,0.2551783003635253,adadebay@umd.edu,false,2023-03-02 22:03,90193.0 secs,Planet Monthly Mosaics,,POINT(34.40926954593081 0.255178300363525),168,Crop +169,169,39.660802154481644,-3.190642695064495,dianafrimpong710@gmail.com,false,2023-03-01 20:43,25.3 secs,Planet Monthly Mosaics,,POINT(39.660802154481644 -3.190642695064495),169,Non-crop +170,170,34.54046798139366,0.6859414994541169,dianafrimpong710@gmail.com,false,2023-03-01 20:43,44.0 secs,Planet Monthly Mosaics,,POINT(34.54046798139366 0.685941499454117),170,Crop +171,171,39.073614246897385,-0.2633723289888229,dianafrimpong710@gmail.com,false,2023-03-01 20:43,8.5 secs,Planet Monthly Mosaics,,POINT(39.073614246897385 -0.263372328988823),171,Non-crop +172,172,38.432919130857414,1.7396830042557871,dianafrimpong710@gmail.com,false,2023-03-01 20:44,15.8 secs,Planet Monthly Mosaics,,POINT(38.432919130857414 1.739683004255787),172,Non-crop +173,173,37.095575381750336,-1.0096069277601003,dianafrimpong710@gmail.com,false,2023-03-01 20:44,7.5 secs,Planet Monthly Mosaics,,POINT(37.095575381750336 -1.0096069277601),173,Crop +174,174,36.65947410135244,-1.1671998161925197,dianafrimpong710@gmail.com,false,2023-03-01 20:45,72.5 secs,Sentinel-2,,POINT(36.65947410135244 -1.16719981619252),174,Crop +175,175,34.45338777219258,-0.7355236000562643,dianafrimpong710@gmail.com,false,2023-03-01 20:46,29.5 secs,Planet Monthly Mosaics,,POINT(34.45338777219258 -0.735523600056264),175,Non-crop +176,176,35.91774080987921,-1.0793531043719524,dianafrimpong710@gmail.com,false,2023-03-01 20:47,60.6 secs,Planet Monthly Mosaics,,POINT(35.91774080987921 -1.079353104371952),176,Crop +177,177,34.72286953140142,0.4356879558642015,dianafrimpong710@gmail.com,false,2023-03-01 20:47,24.7 secs,Planet Monthly Mosaics,,POINT(34.72286953140142 0.435687955864202),177,Non-crop +178,178,35.247612435415604,1.055323072725927,dianafrimpong710@gmail.com,false,2023-03-01 20:47,27.2 secs,Planet Monthly Mosaics,,POINT(35.247612435415604 1.055323072725927),178,Crop +179,179,34.17397768556119,0.41298920794555216,adadebay@umd.edu,false,2023-03-02 22:03,21.8 secs,Mapbox Satellite,,POINT(34.17397768556119 0.412989207945552),179,Non-crop +180,180,36.90739764160443,-0.6045908967298282,dianafrimpong710@gmail.com,false,2023-03-01 20:50,9.2 secs,Planet Monthly Mosaics,,POINT(36.90739764160443 -0.604590896729828),180,Non-crop +181,181,35.09232298964106,-0.24405121259209608,dianafrimpong710@gmail.com,false,2023-03-01 20:50,23.1 secs,Planet Monthly Mosaics,,POINT(35.09232298964106 -0.244051212592096),181,Crop +182,182,34.93320904860172,3.089392923071744,dianafrimpong710@gmail.com,false,2023-03-01 20:50,7.0 secs,Planet Monthly Mosaics,,POINT(34.93320904860172 3.089392923071744),182,Non-crop +183,183,35.386016558505396,0.7061819098528996,dianafrimpong710@gmail.com,false,2023-03-01 20:50,11.0 secs,Planet Monthly Mosaics,,POINT(35.386016558505396 0.7061819098529),183,Crop +184,184,35.44600492019322,0.6737515821752349,dianafrimpong710@gmail.com,false,2023-03-01 20:51,46.5 secs,Planet Monthly Mosaics,,POINT(35.44600492019322 0.673751582175235),184,Crop +185,185,36.83715746471805,-0.3040771550035826,dianafrimpong710@gmail.com,false,2023-03-01 20:52,20.8 secs,Planet Monthly Mosaics,,POINT(36.83715746471805 -0.304077155003583),185,Crop +186,186,35.47235817550058,-1.7747537021443485,dianafrimpong710@gmail.com,false,2023-03-01 20:52,7.2 secs,Planet Monthly Mosaics,,POINT(35.47235817550058 -1.774753702144348),186,Non-crop +187,187,35.9969624367669,-0.05057581718672911,dianafrimpong710@gmail.com,false,2023-03-01 20:52,16.7 secs,Planet Monthly Mosaics,,POINT(35.9969624367669 -0.050575817186729),187,Crop +188,188,35.02138764947741,0.9068522061088463,dianafrimpong710@gmail.com,false,2023-03-01 20:53,35.2 secs,Planet Monthly Mosaics,,POINT(35.02138764947741 0.906852206108846),188,Non-crop +189,189,37.28290916070205,-0.5670627725341534,adadebay@umd.edu,false,2023-03-02 22:06,158.6 secs,Mapbox Satellite,,POINT(37.28290916070205 -0.567062772534153),189,Non-crop +190,190,36.27531089330441,1.6312781626526536,dianafrimpong710@gmail.com,false,2023-03-01 20:55,20.5 secs,Planet Monthly Mosaics,,POINT(36.27531089330441 1.631278162652654),190,Non-crop +191,191,38.18613982888146,2.3095627412092896,dianafrimpong710@gmail.com,false,2023-03-01 20:55,10.1 secs,Planet Monthly Mosaics,,POINT(38.18613982888146 2.30956274120929),191,Non-crop +192,192,38.82823907086845,2.961020075278439,dianafrimpong710@gmail.com,false,2023-03-01 20:56,22.3 secs,Planet Monthly Mosaics,,POINT(38.82823907086845 2.961020075278439),192,Non-crop +193,193,35.756354105431015,-0.25214860600974354,dianafrimpong710@gmail.com,false,2023-03-01 20:56,26.0 secs,Planet Monthly Mosaics,,POINT(35.756354105431015 -0.252148606009744),193,Crop +194,194,34.05911892769959,-0.7668905312894166,dianafrimpong710@gmail.com,false,2023-03-01 20:56,8.5 secs,Planet Monthly Mosaics,,POINT(34.05911892769959 -0.766890531289417),194,Non-crop +195,195,37.434398228197466,-1.4124646108398355,adadebay@umd.edu,false,2023-03-02 22:06,29.6 secs,Planet Monthly Mosaics,,POINT(37.434398228197466 -1.412464610839836),195,Crop +196,196,34.24813006085403,-0.18754586249713845,dianafrimpong710@gmail.com,false,2023-03-01 20:58,15.6 secs,Planet Monthly Mosaics,,POINT(34.24813006085403 -0.187545862497138),196,Non-crop +197,197,39.38163103579628,-0.5870464083569815,dianafrimpong710@gmail.com,false,2023-03-01 20:58,6.9 secs,Planet Monthly Mosaics,,POINT(39.38163103579628 -0.587046408356982),197,Non-crop +198,198,34.407612059311866,4.020736333189546,dianafrimpong710@gmail.com,false,2023-03-01 20:58,6.4 secs,Planet Monthly Mosaics,,POINT(34.407612059311866 4.020736333189546),198,Non-crop +199,199,34.5106962080211,-0.1159075048798257,dianafrimpong710@gmail.com,false,2023-03-01 20:59,77.3 secs,Planet Monthly Mosaics,,POINT(34.5106962080211 -0.115907504879826),199,Crop +200,200,39.699070498363014,0.8268795354015316,dianafrimpong710@gmail.com,false,2023-03-01 20:59,6.5 secs,Planet Monthly Mosaics,,POINT(39.699070498363014 0.826879535401532),200,Non-crop +201,201,35.26606385102699,-0.8488962562008477,dianafrimpong710@gmail.com,false,2023-03-01 21:01,71.0 secs,Sentinel-2,,POINT(35.26606385102699 -0.848896256200848),201,Non-crop +202,202,37.00569584470588,-0.4607649473863657,dianafrimpong710@gmail.com,false,2023-03-01 21:02,69.1 secs,Planet Monthly Mosaics,,POINT(37.00569584470588 -0.460764947386366),202,Crop +203,203,34.681833787046784,-0.0017856800434060153,dianafrimpong710@gmail.com,false,2023-03-01 21:03,98.8 secs,Planet Monthly Mosaics,,POINT(34.681833787046784 -0.001785680043406),203,Crop +204,204,35.00322143113498,-0.37269709187609834,adadebay@umd.edu,false,2023-03-02 22:07,29.3 secs,Mapbox Satellite,,POINT(35.00322143113498 -0.372697091876098),204,Non-crop +205,205,35.09439243911183,0.4304058574197333,dianafrimpong710@gmail.com,false,2023-03-01 21:06,47.2 secs,Planet Monthly Mosaics,,POINT(35.09439243911183 0.430405857419733),205,Crop +206,206,35.13586471270375,3.8666126635096743,dianafrimpong710@gmail.com,false,2023-03-01 21:06,13.0 secs,Planet Monthly Mosaics,,POINT(35.13586471270375 3.866612663509674),206,Non-crop +207,207,38.15212804664146,3.3677157950076575,dianafrimpong710@gmail.com,false,2023-03-01 21:06,8.9 secs,Planet Monthly Mosaics,,POINT(38.15212804664146 3.367715795007658),207,Non-crop +208,208,38.286578480177624,-3.580862928287623,dianafrimpong710@gmail.com,false,2023-03-01 21:09,156.5 secs,Planet Monthly Mosaics,,POINT(38.286578480177624 -3.580862928287623),208,Non-crop +209,209,39.404093137783995,2.45152108353291,dianafrimpong710@gmail.com,false,2023-03-01 21:09,22.4 secs,Planet Monthly Mosaics,,POINT(39.404093137783995 2.45152108353291),209,Non-crop +210,210,35.065711700959476,2.547656684794862,dianafrimpong710@gmail.com,false,2023-03-01 21:09,7.5 secs,Planet Monthly Mosaics,,POINT(35.065711700959476 2.547656684794862),210,Non-crop +211,211,35.947769272904495,0.20415397808837918,dianafrimpong710@gmail.com,false,2023-03-01 21:10,46.9 secs,Planet Monthly Mosaics,,POINT(35.947769272904495 0.204153978088379),211,Non-crop +212,212,37.91801786296307,-2.1375884435327785,dianafrimpong710@gmail.com,false,2023-03-01 21:12,107.0 secs,Planet Monthly Mosaics,,POINT(37.91801786296307 -2.137588443532778),212,Non-crop +213,213,36.88669703425683,0.42706564857485335,dianafrimpong710@gmail.com,false,2023-03-01 21:12,17.0 secs,Planet Monthly Mosaics,,POINT(36.88669703425683 0.427065648574853),213,Non-crop +214,214,34.068160039789994,-0.6537499847545581,dianafrimpong710@gmail.com,false,2023-03-01 21:12,19.7 secs,Planet Monthly Mosaics,,POINT(34.068160039789994 -0.653749984754558),214,Crop +215,215,37.99915424752388,0.06625355062429386,dianafrimpong710@gmail.com,false,2023-03-01 21:15,136.3 secs,Planet Monthly Mosaics,,POINT(37.99915424752388 0.066253550624294),215,Non-crop +216,216,38.23840190779963,1.666609091145675,dianafrimpong710@gmail.com,false,2023-03-01 21:15,13.1 secs,Planet Monthly Mosaics,,POINT(38.23840190779963 1.666609091145675),216,Non-crop +217,217,38.09419036742432,2.9609319691321545,dianafrimpong710@gmail.com,false,2023-03-01 21:15,15.5 secs,Planet Monthly Mosaics,,POINT(38.09419036742432 2.960931969132154),217,Non-crop +218,218,40.11268741129131,0.07066709430285514,dianafrimpong710@gmail.com,false,2023-03-01 21:15,6.9 secs,Planet Monthly Mosaics,,POINT(40.11268741129131 0.070667094302855),218,Non-crop +219,219,35.33297997639196,-0.8617340640007177,dianafrimpong710@gmail.com,false,2023-03-01 21:16,33.4 secs,Planet Monthly Mosaics,,POINT(35.33297997639196 -0.861734064000718),219,Crop +220,220,35.85807814705425,4.10000582662628,dianafrimpong710@gmail.com,false,2023-03-01 21:16,6.9 secs,Planet Monthly Mosaics,,POINT(35.85807814705425 4.10000582662628),220,Non-crop +221,221,34.5729975776531,-0.6466759300640075,dianafrimpong710@gmail.com,false,2023-03-01 21:16,7.1 secs,Planet Monthly Mosaics,,POINT(34.5729975776531 -0.646675930064008),221,Crop +222,222,40.546244593875265,1.1821031768260064,dianafrimpong710@gmail.com,false,2023-03-01 21:16,5.5 secs,Planet Monthly Mosaics,,POINT(40.546244593875265 1.182103176826006),222,Non-crop +223,223,37.548413950439276,-0.41160735763497225,dianafrimpong710@gmail.com,false,2023-03-01 21:19,146.9 secs,Planet Monthly Mosaics,,POINT(37.548413950439276 -0.411607357634972),223,Non-crop +224,224,35.75968549281819,-0.9621064827168123,dianafrimpong710@gmail.com,false,2023-03-01 21:19,15.7 secs,Planet Monthly Mosaics,,POINT(35.75968549281819 -0.962106482716812),224,Crop +225,225,35.074744972874576,-0.7925917814780629,dianafrimpong710@gmail.com,false,2023-03-01 21:19,30.6 secs,Planet Monthly Mosaics,,POINT(35.074744972874576 -0.792591781478063),225,Crop +226,226,34.46024019008391,3.53618931300853,dianafrimpong710@gmail.com,false,2023-03-01 21:20,6.3 secs,Planet Monthly Mosaics,,POINT(34.46024019008391 3.53618931300853),226,Non-crop +227,227,38.83677570935286,2.824285402842024,dianafrimpong710@gmail.com,false,2023-03-01 21:20,6.1 secs,Planet Monthly Mosaics,,POINT(38.83677570935286 2.824285402842024),227,Non-crop +228,228,37.397368184916004,-0.6140394095524093,dianafrimpong710@gmail.com,false,2023-03-01 21:20,26.3 secs,Planet Monthly Mosaics,,POINT(37.397368184916004 -0.614039409552409),228,Crop +229,229,35.762616150651965,-1.8264944601622872,dianafrimpong710@gmail.com,false,2023-03-01 21:20,18.8 secs,Planet Monthly Mosaics,,POINT(35.762616150651965 -1.826494460162287),229,Non-crop +230,230,36.54848043845705,-1.000772776636237,dianafrimpong710@gmail.com,false,2023-03-01 21:22,72.8 secs,Sentinel-2,,POINT(36.54848043845705 -1.000772776636237),230,Crop +231,231,36.02608988189172,3.627360664812145,dianafrimpong710@gmail.com,false,2023-03-01 21:22,18.8 secs,Planet Monthly Mosaics,,POINT(36.02608988189172 3.627360664812145),231,Non-crop +232,232,33.98152361411099,-0.47111917194585895,dianafrimpong710@gmail.com,false,2023-03-01 21:23,42.5 secs,Planet Monthly Mosaics,,POINT(33.98152361411099 -0.471119171945859),232,Non-crop +233,233,36.08813405957186,-0.941069865990557,dianafrimpong710@gmail.com,false,2023-03-01 21:23,12.4 secs,Planet Monthly Mosaics,,POINT(36.08813405957186 -0.941069865990557),233,Crop +234,234,37.60598532487981,-0.717185806924993,dianafrimpong710@gmail.com,false,2023-03-01 21:24,40.2 secs,Planet Monthly Mosaics,,POINT(37.60598532487981 -0.717185806924993),234,Non-crop +235,235,34.46747779967232,0.8305382197080086,dianafrimpong710@gmail.com,false,2023-03-01 21:24,35.8 secs,Planet Monthly Mosaics,,POINT(34.46747779967232 0.830538219708009),235,Crop +236,236,34.840773987663475,-0.36183332301270066,dianafrimpong710@gmail.com,false,2023-03-01 21:25,28.0 secs,Planet Monthly Mosaics,,POINT(34.840773987663475 -0.361833323012701),236,Non-crop +237,237,37.41092767037453,-2.625588354308176,dianafrimpong710@gmail.com,false,2023-03-01 21:25,5.5 secs,Planet Monthly Mosaics,,POINT(37.41092767037453 -2.625588354308176),237,Non-crop +238,238,34.5067667042273,4.001073110839722,dianafrimpong710@gmail.com,false,2023-03-01 21:25,6.0 secs,Planet Monthly Mosaics,,POINT(34.5067667042273 4.001073110839722),238,Non-crop +239,239,37.67487855659351,-0.49020904050426295,dianafrimpong710@gmail.com,false,2023-03-01 21:25,18.9 secs,Planet Monthly Mosaics,,POINT(37.67487855659351 -0.490209040504263),239,Crop +240,240,36.81888775078636,1.9849372691074236,dianafrimpong710@gmail.com,false,2023-03-01 21:25,6.7 secs,Planet Monthly Mosaics,,POINT(36.81888775078636 1.984937269107424),240,Non-crop +241,241,35.24994048248167,-1.0513444774351794,dianafrimpong710@gmail.com,false,2023-03-01 21:25,10.6 secs,Planet Monthly Mosaics,,POINT(35.24994048248167 -1.051344477435179),241,Crop +242,242,36.38636671973139,-0.5930179628924186,dianafrimpong710@gmail.com,false,2023-03-01 21:26,15.0 secs,Planet Monthly Mosaics,,POINT(36.38636671973139 -0.593017962892419),242,Non-crop +243,243,40.907140441399896,0.6303587257539014,dianafrimpong710@gmail.com,false,2023-03-01 21:26,15.0 secs,Planet Monthly Mosaics,,POINT(40.907140441399896 0.630358725753901),243,Non-crop +244,244,34.96987624938218,1.1655669652004106,dianafrimpong710@gmail.com,false,2023-03-01 21:26,12.3 secs,Planet Monthly Mosaics,,POINT(34.96987624938218 1.165566965200411),244,Crop +245,245,35.35634367392324,1.4582707365296375,dianafrimpong710@gmail.com,false,2023-03-01 21:26,12.9 secs,Planet Monthly Mosaics,,POINT(35.35634367392324 1.458270736529638),245,Non-crop +246,246,40.546097997935725,-0.07612710307294997,dianafrimpong710@gmail.com,false,2023-03-01 21:26,5.6 secs,Planet Monthly Mosaics,,POINT(40.546097997935725 -0.07612710307295),246,Non-crop +247,247,37.16390525162651,-0.7420742630040361,dianafrimpong710@gmail.com,false,2023-03-01 21:27,22.3 secs,Planet Monthly Mosaics,,POINT(37.16390525162651 -0.742074263004036),247,Crop +248,248,34.45068841495048,0.2338980943932806,dianafrimpong710@gmail.com,false,2023-03-01 21:27,20.6 secs,Planet Monthly Mosaics,,POINT(34.45068841495048 0.233898094393281),248,Crop +249,249,37.498286295079225,0.9862352131017155,dianafrimpong710@gmail.com,false,2023-03-01 21:27,16.0 secs,Planet Monthly Mosaics,,POINT(37.498286295079225 0.986235213101716),249,Non-crop +250,250,34.73415295793003,0.018438781875223693,adadebay@umd.edu,false,2023-03-02 22:07,6.4 secs,Mapbox Satellite,,POINT(34.73415295793003 0.018438781875224),250,Non-crop +251,251,36.85858969388467,-0.4686485504993926,adadebay@umd.edu,false,2023-03-02 22:07,8.6 secs,Mapbox Satellite,,POINT(36.85858969388467 -0.468648550499393),251,Crop +252,252,35.26890826320225,-0.6564145540663221,adadebay@umd.edu,false,2023-03-02 22:07,5.8 secs,Mapbox Satellite,,POINT(35.26890826320225 -0.656414554066322),252,Crop +253,253,40.525409290323495,2.6319993375838955,adadebay@umd.edu,false,2023-03-02 22:07,4.9 secs,Mapbox Satellite,,POINT(40.525409290323495 2.631999337583896),253,Non-crop +254,254,34.92265893447306,0.5353286427591524,adadebay@umd.edu,false,2023-03-02 22:07,5.6 secs,Mapbox Satellite,,POINT(34.92265893447306 0.535328642759152),254,Non-crop +255,255,36.90455311386702,1.1977744701246156,adadebay@umd.edu,false,2023-03-02 22:07,4.7 secs,Mapbox Satellite,,POINT(36.90455311386702 1.197774470124616),255,Non-crop +256,256,34.5180586260015,0.7789264000463355,adadebay@umd.edu,false,2023-03-02 22:08,16.2 secs,Mapbox Satellite,,POINT(34.5180586260015 0.778926400046336),256,Non-crop +257,257,36.058209274389014,-1.3463982887488246,adadebay@umd.edu,false,2023-03-02 22:08,5.3 secs,Mapbox Satellite,,POINT(36.058209274389014 -1.346398288748825),257,Non-crop +258,258,39.395468711706776,2.0383453012130155,adadebay@umd.edu,false,2023-03-02 22:08,4.9 secs,Mapbox Satellite,,POINT(39.395468711706776 2.038345301213016),258,Non-crop +259,259,39.05025716334389,1.2727642254608016,adadebay@umd.edu,false,2023-03-02 22:08,5.3 secs,Mapbox Satellite,,POINT(39.05025716334389 1.272764225460802),259,Non-crop +260,260,40.78878641079977,0.947351946589384,adadebay@umd.edu,false,2023-03-02 22:08,4.7 secs,Mapbox Satellite,,POINT(40.78878641079977 0.947351946589384),260,Non-crop +261,261,39.46290672282505,-0.013057674497699916,adadebay@umd.edu,false,2023-03-02 22:08,5.0 secs,Mapbox Satellite,,POINT(39.46290672282505 -0.0130576744977),261,Non-crop +262,262,35.19685108842834,3.9040331699588258,adadebay@umd.edu,false,2023-03-02 22:08,11.5 secs,Mapbox Satellite,,POINT(35.19685108842834 3.904033169958826),262,Non-crop +263,263,37.299210366889746,-0.7882590697784982,adadebay@umd.edu,false,2023-03-03 17:29,23.2 secs,Mapbox Satellite,,POINT(37.299210366889746 -0.788259069778498),263,Non-crop +264,264,35.134539162750926,1.0904367967402915,adadebay@umd.edu,false,2023-03-03 17:30,24.2 secs,Mapbox Satellite,,POINT(35.134539162750926 1.090436796740292),264,Crop +265,265,38.19300016102455,-1.6485412210153514,adadebay@umd.edu,false,2023-03-03 17:51,1271.7 secs,Planet Monthly Mosaics,,POINT(38.19300016102455 -1.648541221015351),265,Non-crop +266,266,37.73860020267111,-3.331557513616808,adadebay@umd.edu,false,2023-03-03 17:51,12.5 secs,Mapbox Satellite,,POINT(37.73860020267111 -3.331557513616808),266,Non-crop +267,267,36.183154340115976,2.7001405183904446,adadebay@umd.edu,false,2023-03-03 17:51,5.2 secs,Mapbox Satellite,,POINT(36.183154340115976 2.700140518390445),267,Non-crop +268,268,39.9003149820409,3.4102972857651417,adadebay@umd.edu,false,2023-03-03 17:51,5.0 secs,Mapbox Satellite,,POINT(39.9003149820409 3.410297285765142),268,Non-crop +269,269,36.89256033951802,-1.198551976200557,adadebay@umd.edu,false,2023-03-03 17:51,10.6 secs,Mapbox Satellite,,POINT(36.89256033951802 -1.198551976200557),269,Non-crop +270,270,34.111742992464094,-0.03294774771910352,adadebay@umd.edu,false,2023-03-03 17:51,8.1 secs,Mapbox Satellite,,POINT(34.111742992464094 -0.032947747719104),270,Non-crop +271,271,38.55271191716017,-1.5870930361786544,adadebay@umd.edu,false,2023-03-03 17:52,4.8 secs,Mapbox Satellite,,POINT(38.55271191716017 -1.587093036178654),271,Non-crop +272,272,35.79147363846596,1.8930771879554555,adadebay@umd.edu,false,2023-03-03 17:52,9.3 secs,Mapbox Satellite,,POINT(35.79147363846596 1.893077187955456),272,Non-crop +273,273,35.0802486455078,0.7325502982864748,adadebay@umd.edu,false,2023-03-03 17:52,12.5 secs,Mapbox Satellite,,POINT(35.0802486455078 0.732550298286475),273,Crop +274,274,36.66787862075995,-0.3370048147857069,adadebay@umd.edu,false,2023-03-03 17:52,7.4 secs,Mapbox Satellite,,POINT(36.66787862075995 -0.337004814785707),274,Non-crop +275,275,35.382128847748994,-0.8625910503550166,adadebay@umd.edu,false,2023-03-03 17:53,32.1 secs,Mapbox Satellite,,POINT(35.382128847748994 -0.862591050355017),275,Non-crop +276,276,37.34840090595225,0.23193127521008539,adadebay@umd.edu,false,2023-03-03 17:53,5.1 secs,Mapbox Satellite,,POINT(37.34840090595225 0.231931275210085),276,Non-crop +277,277,37.52952524833004,-2.724401129435582,adadebay@umd.edu,false,2023-03-03 17:53,5.8 secs,Mapbox Satellite,,POINT(37.52952524833004 -2.724401129435582),277,Non-crop +278,278,34.30386139976155,0.2951159530038938,adadebay@umd.edu,false,2023-03-03 17:53,35.7 secs,Mapbox Satellite,,POINT(34.30386139976155 0.295115953003894),278,Non-crop +279,279,35.664840363056655,1.690372204559844,dianafrimpong710@gmail.com,false,2023-03-03 21:31,41.4 secs,Planet Monthly Mosaics,,POINT(35.664840363056655 1.690372204559844),279,Non-crop +280,280,39.19739507190026,-1.2216252242734178,dianafrimpong710@gmail.com,false,2023-03-03 21:32,21.0 secs,Planet Monthly Mosaics,,POINT(39.19739507190026 -1.221625224273418),280,Non-crop +281,281,35.184928340727154,1.8744824920807754,dianafrimpong710@gmail.com,false,2023-03-03 21:32,22.3 secs,Planet Monthly Mosaics,,POINT(35.184928340727154 1.874482492080775),281,Non-crop +282,282,37.78857700274938,-3.3771286904231905,dianafrimpong710@gmail.com,false,2023-03-03 21:33,54.1 secs,Planet Monthly Mosaics,,POINT(37.78857700274938 -3.37712869042319),282,Non-crop +283,283,36.68502729602952,2.514325852545885,dianafrimpong710@gmail.com,false,2023-03-03 21:33,22.1 secs,Planet Monthly Mosaics,,POINT(36.68502729602952 2.514325852545885),283,Non-crop +284,284,35.237040051019214,1.6524385569289506,dianafrimpong710@gmail.com,false,2023-03-03 21:34,55.6 secs,Planet Monthly Mosaics,,POINT(35.237040051019214 1.652438556928951),284,Crop +285,285,36.53526300735865,-0.4683613786490718,dianafrimpong710@gmail.com,false,2023-03-03 21:35,60.9 secs,Planet Monthly Mosaics,,POINT(36.53526300735865 -0.468361378649072),285,Crop +286,286,38.0760525856689,-0.11913545569567449,dianafrimpong710@gmail.com,false,2023-03-03 21:37,85.5 secs,Planet Monthly Mosaics,,POINT(38.0760525856689 -0.119135455695674),286,Non-crop +287,287,34.35644028067466,0.5608866070711257,dianafrimpong710@gmail.com,false,2023-03-03 21:37,29.1 secs,Planet Monthly Mosaics,,POINT(34.35644028067466 0.560886607071126),287,Crop +288,288,38.65484794845974,-0.4860956688299213,dianafrimpong710@gmail.com,false,2023-03-03 21:38,36.8 secs,Planet Monthly Mosaics,,POINT(38.65484794845974 -0.486095668829921),288,Non-crop +289,289,35.93734608846335,-1.057140931452576,dianafrimpong710@gmail.com,false,2023-03-03 21:38,30.5 secs,Planet Monthly Mosaics,,POINT(35.93734608846335 -1.057140931452576),289,Non-crop +290,290,37.80582391504641,3.8350499412603556,dianafrimpong710@gmail.com,false,2023-03-03 21:39,25.1 secs,Planet Monthly Mosaics,,POINT(37.80582391504641 3.835049941260356),290,Non-crop +291,291,35.25737937467433,-0.8666069188032641,dianafrimpong710@gmail.com,false,2023-03-03 21:39,35.5 secs,Planet Monthly Mosaics,,POINT(35.25737937467433 -0.866606918803264),291,Non-crop +292,292,34.35788383131346,0.19694288433069554,dianafrimpong710@gmail.com,false,2023-03-03 21:45,345.2 secs,Planet Monthly Mosaics,,POINT(34.35788383131346 0.196942884330696),292,Crop +293,293,34.818940898978965,-1.2688823327413155,dianafrimpong710@gmail.com,false,2023-03-03 21:45,22.4 secs,Planet Monthly Mosaics,,POINT(34.818940898978965 -1.268882332741316),293,Non-crop +294,294,34.93880589858098,2.3229353279690117,dianafrimpong710@gmail.com,false,2023-03-03 21:45,10.0 secs,Planet Monthly Mosaics,,POINT(34.93880589858098 2.322935327969012),294,Non-crop +295,295,39.054542526331986,0.0026259847551115003,dianafrimpong710@gmail.com,false,2023-03-03 21:46,11.7 secs,Planet Monthly Mosaics,,POINT(39.054542526331986 0.002625984755112),295,Non-crop +296,296,36.07763478509327,-1.1708826042522547,dianafrimpong710@gmail.com,false,2023-03-03 21:47,98.0 secs,Planet Monthly Mosaics,,POINT(36.07763478509327 -1.170882604252255),296,Crop +297,297,34.90669887109912,0.7293370301766702,dianafrimpong710@gmail.com,false,2023-03-03 21:53,375.6 secs,Planet Monthly Mosaics,,POINT(34.90669887109912 0.72933703017667),297,Crop +298,298,38.79853398079912,0.9642160077813193,dianafrimpong710@gmail.com,false,2023-03-03 21:54,10.7 secs,Planet Monthly Mosaics,,POINT(38.79853398079912 0.964216007781319),298,Non-crop +299,299,37.553857774558885,-2.9026633735747365,dianafrimpong710@gmail.com,false,2023-03-03 21:54,36.8 secs,Planet Monthly Mosaics,,POINT(37.553857774558885 -2.902663373574736),299,Crop +300,300,38.01347608722678,2.900570657102886,dianafrimpong710@gmail.com,false,2023-03-03 21:55,40.1 secs,Planet Monthly Mosaics,,POINT(38.01347608722678 2.900570657102886),300,Non-crop +301,301,37.132692772610476,0.39776039039835437,dianafrimpong710@gmail.com,false,2023-03-03 21:55,19.2 secs,Planet Monthly Mosaics,,POINT(37.132692772610476 0.397760390398354),301,Non-crop +302,302,37.9392642060917,2.9892719733950126,dianafrimpong710@gmail.com,false,2023-03-03 21:55,9.9 secs,Planet Monthly Mosaics,,POINT(37.9392642060917 2.989271973395013),302,Non-crop +303,303,34.51506919888127,0.1202453794900851,dianafrimpong710@gmail.com,false,2023-03-03 21:55,9.8 secs,Planet Monthly Mosaics,,POINT(34.51506919888127 0.120245379490085),303,Crop +304,304,39.63767229599631,-0.12049561541535113,dianafrimpong710@gmail.com,false,2023-03-03 21:56,10.8 secs,Planet Monthly Mosaics,,POINT(39.63767229599631 -0.120495615415351),304,Non-crop +305,305,40.485326212115005,3.093570121762987,dianafrimpong710@gmail.com,false,2023-03-03 21:56,9.1 secs,Planet Monthly Mosaics,,POINT(40.485326212115005 3.093570121762987),305,Non-crop +306,306,36.85239673523007,-0.05853659594733882,dianafrimpong710@gmail.com,false,2023-03-03 21:56,22.1 secs,Planet Monthly Mosaics,,POINT(36.85239673523007 -0.058536595947339),306,Non-crop +307,307,37.5763483597854,-0.9823710756791123,dianafrimpong710@gmail.com,false,2023-03-03 21:57,30.6 secs,Planet Monthly Mosaics,,POINT(37.5763483597854 -0.982371075679112),307,Non-crop +308,308,39.55791935681194,1.5349026436085218,dianafrimpong710@gmail.com,false,2023-03-03 21:57,11.2 secs,Planet Monthly Mosaics,,POINT(39.55791935681194 1.534902643608522),308,Non-crop +309,309,37.699949750619936,0.811948379073906,dianafrimpong710@gmail.com,false,2023-03-03 22:08,682.9 secs,Planet Monthly Mosaics,,POINT(37.699949750619936 0.811948379073906),309,Non-crop +310,310,37.61422072618653,-2.0872413462062473,adadebay@umd.edu,false,2023-03-06 13:02,11.2 secs,Mapbox Satellite,,POINT(37.61422072618653 -2.087241346206247),310,Non-crop +311,311,41.07513171921359,3.1888676288992936,adadebay@umd.edu,false,2023-03-06 13:02,5.1 secs,Mapbox Satellite,,POINT(41.07513171921359 3.188867628899294),311,Non-crop +312,312,38.15815408847578,-0.8111453505853832,adadebay@umd.edu,false,2023-03-06 13:02,6.5 secs,Mapbox Satellite,,POINT(38.15815408847578 -0.811145350585383),312,Non-crop +313,313,34.736270179275344,-0.5624127830726717,adadebay@umd.edu,false,2023-03-06 13:02,6.3 secs,Mapbox Satellite,,POINT(34.736270179275344 -0.562412783072672),313,Crop +314,314,35.375810528396606,-0.21120699908951038,adadebay@umd.edu,false,2023-03-06 13:03,8.4 secs,Mapbox Satellite,,POINT(35.375810528396606 -0.21120699908951),314,Crop +315,315,37.21904951713753,-0.7068700226033653,adadebay@umd.edu,false,2023-03-06 13:03,7.0 secs,Mapbox Satellite,,POINT(37.21904951713753 -0.706870022603365),315,Non-crop +316,316,39.83400661063378,-1.68890740390545,adadebay@umd.edu,false,2023-03-06 13:03,4.7 secs,Mapbox Satellite,,POINT(39.83400661063378 -1.68890740390545),316,Non-crop +317,317,34.191058880953285,-0.9215307463490227,adadebay@umd.edu,false,2023-03-06 13:03,5.5 secs,Mapbox Satellite,,POINT(34.191058880953285 -0.921530746349023),317,Crop +318,318,34.49184864640101,-0.7460763000252552,adadebay@umd.edu,false,2023-03-06 13:03,6.1 secs,Mapbox Satellite,,POINT(34.49184864640101 -0.746076300025255),318,Crop +319,319,37.651127920351016,-0.6203277536564297,adadebay@umd.edu,false,2023-03-06 13:03,13.7 secs,Mapbox Satellite,,POINT(37.651127920351016 -0.62032775365643),319,Non-crop +320,320,38.75780790144454,-0.6667358112308936,adadebay@umd.edu,false,2023-03-06 13:04,28.5 secs,Mapbox Satellite,,POINT(38.75780790144454 -0.666735811230894),320,Non-crop +321,321,34.94783457049091,-0.11246745761080304,adadebay@umd.edu,false,2023-03-06 13:04,19.6 secs,Mapbox Satellite,,POINT(34.94783457049091 -0.112467457610803),321,Non-crop +322,322,39.229061071429356,-0.666375893413772,adadebay@umd.edu,false,2023-03-06 13:04,5.6 secs,Mapbox Satellite,,POINT(39.229061071429356 -0.666375893413772),322,Non-crop +323,323,38.88788044369188,-4.24884778504972,adadebay@umd.edu,false,2023-03-06 13:04,6.5 secs,Mapbox Satellite,,POINT(38.88788044369188 -4.24884778504972),323,Non-crop +324,324,36.71627653401663,2.0002891318295455,adadebay@umd.edu,false,2023-03-06 13:04,7.1 secs,Mapbox Satellite,,POINT(36.71627653401663 2.000289131829546),324,Non-crop +325,325,35.99051353043517,-1.0041755970886126,adadebay@umd.edu,false,2023-03-06 13:04,5.9 secs,Mapbox Satellite,,POINT(35.99051353043517 -1.004175597088613),325,Non-crop +326,326,34.693116889906136,0.7036994275098215,adadebay@umd.edu,false,2023-03-06 13:05,6.0 secs,Mapbox Satellite,,POINT(34.693116889906136 0.703699427509822),326,Crop +327,327,36.80110616930354,-1.3991582831341616,adadebay@umd.edu,false,2023-03-06 13:05,12.0 secs,Mapbox Satellite,,POINT(36.80110616930354 -1.399158283134162),327,Non-crop +328,328,34.359713263958206,3.7495077872591676,adadebay@umd.edu,false,2023-03-06 13:05,5.3 secs,Mapbox Satellite,,POINT(34.359713263958206 3.749507787259168),328,Non-crop +329,329,34.89872798832474,3.7711850926712276,adadebay@umd.edu,false,2023-03-06 13:05,5.8 secs,Mapbox Satellite,,POINT(34.89872798832474 3.771185092671228),329,Non-crop +330,330,34.75823814690865,0.768390378516022,adadebay@umd.edu,false,2023-03-06 13:05,10.5 secs,Mapbox Satellite,,POINT(34.75823814690865 0.768390378516022),330,Non-crop +331,331,35.03043307620979,-0.435833843994421,adadebay@umd.edu,false,2023-03-06 13:05,6.6 secs,Mapbox Satellite,,POINT(35.03043307620979 -0.435833843994421),331,Crop +332,332,39.91725992960783,-2.536230581743613,adadebay@umd.edu,false,2023-03-06 13:07,76.8 secs,Sentinel-2,,POINT(39.91725992960783 -2.536230581743613),332,Non-crop +333,333,38.04110617648035,2.134810668583166,adadebay@umd.edu,false,2023-03-06 13:07,31.1 secs,Mapbox Satellite,,POINT(38.04110617648035 2.134810668583166),333,Non-crop +334,334,38.616261925366786,2.5167846397007954,adadebay@umd.edu,false,2023-03-06 13:07,6.2 secs,Mapbox Satellite,,POINT(38.616261925366786 2.516784639700795),334,Non-crop +335,335,36.910665570255915,-1.1618142066887174,adadebay@umd.edu,false,2023-03-06 13:08,77.1 secs,Sentinel-2,,POINT(36.910665570255915 -1.161814206688717),335,Crop +336,336,36.42768875867901,2.649630024893417,adadebay@umd.edu,false,2023-03-06 13:09,14.0 secs,Sentinel-2,,POINT(36.42768875867901 2.649630024893417),336,Non-crop +337,337,37.41839115014418,-1.0816838435200997,adadebay@umd.edu,false,2023-03-06 13:10,77.2 secs,Planet Monthly Mosaics,,POINT(37.41839115014418 -1.0816838435201),337,Crop +338,338,37.10693932494706,-0.6350245887261098,adadebay@umd.edu,false,2023-03-06 13:10,19.2 secs,Mapbox Satellite,,POINT(37.10693932494706 -0.63502458872611),338,Non-crop +339,339,36.984429661995755,3.311675194970429,adadebay@umd.edu,false,2023-03-06 13:10,5.4 secs,Mapbox Satellite,,POINT(36.984429661995755 3.311675194970429),339,Non-crop +340,340,36.144412583364996,0.004963373055491912,adadebay@umd.edu,false,2023-03-06 13:11,29.8 secs,Mapbox Satellite,,POINT(36.144412583364996 0.004963373055492),340,Non-crop +341,341,36.70078650447418,3.309703935432483,adadebay@umd.edu,false,2023-03-06 13:11,5.5 secs,Mapbox Satellite,,POINT(36.70078650447418 3.309703935432483),341,Non-crop +342,342,38.75161279448816,2.378729348687549,adadebay@umd.edu,false,2023-03-06 13:11,5.9 secs,Mapbox Satellite,,POINT(38.75161279448816 2.378729348687549),342,Non-crop +343,343,40.06707636849588,3.45275268077823,adadebay@umd.edu,false,2023-03-06 13:11,4.8 secs,Mapbox Satellite,,POINT(40.06707636849588 3.45275268077823),343,Non-crop +344,344,36.56210500652624,-1.0837246415363992,adadebay@umd.edu,false,2023-03-06 13:11,4.8 secs,Mapbox Satellite,,POINT(36.56210500652624 -1.083724641536399),344,Non-crop +345,345,34.336041577902456,0.7166491969812425,adadebay@umd.edu,false,2023-03-06 13:12,23.0 secs,Planet Monthly Mosaics,,POINT(34.336041577902456 0.716649196981242),345,Crop +346,346,39.598853154828326,2.8858817878871452,adadebay@umd.edu,false,2023-03-06 13:12,10.0 secs,Mapbox Satellite,,POINT(39.598853154828326 2.885881787887145),346,Non-crop +347,347,36.43918846011223,1.440504544935712,adadebay@umd.edu,false,2023-03-06 13:12,5.1 secs,Mapbox Satellite,,POINT(36.43918846011223 1.440504544935712),347,Non-crop +348,348,35.54210939603061,-1.327991441743508,adadebay@umd.edu,false,2023-03-06 13:12,5.7 secs,Mapbox Satellite,,POINT(35.54210939603061 -1.327991441743508),348,Non-crop +349,349,34.24661018198986,3.7788091335579623,adadebay@umd.edu,false,2023-03-06 13:12,5.0 secs,Mapbox Satellite,,POINT(34.24661018198986 3.778809133557962),349,Non-crop +350,350,38.155177833382425,-0.9094747900561307,adadebay@umd.edu,false,2023-03-06 13:12,6.9 secs,Mapbox Satellite,,POINT(38.155177833382425 -0.909474790056131),350,Non-crop +351,351,37.72392326099627,-2.8587133707172647,adadebay@umd.edu,false,2023-03-06 13:12,7.6 secs,Mapbox Satellite,,POINT(37.72392326099627 -2.858713370717265),351,Non-crop +352,352,39.185711371300165,0.11214360263653626,adadebay@umd.edu,false,2023-03-06 13:12,5.6 secs,Mapbox Satellite,,POINT(39.185711371300165 0.112143602636536),352,Non-crop +353,353,36.81517663408579,0.06603508799998077,adadebay@umd.edu,false,2023-03-06 13:13,5.3 secs,Mapbox Satellite,,POINT(36.81517663408579 0.066035087999981),353,Non-crop +354,354,39.73857946305931,1.3005086068260963,adadebay@umd.edu,false,2023-03-06 13:13,5.9 secs,Mapbox Satellite,,POINT(39.73857946305931 1.300508606826096),354,Non-crop +355,355,34.89094893452659,0.5432189512967784,adadebay@umd.edu,false,2023-03-06 13:13,6.2 secs,Mapbox Satellite,,POINT(34.89094893452659 0.543218951296778),355,Crop +356,356,35.34730497147133,1.8974096389644381,adadebay@umd.edu,false,2023-03-06 13:13,5.8 secs,Mapbox Satellite,,POINT(35.34730497147133 1.897409638964438),356,Non-crop +357,357,39.4560023573566,-3.7589481349304537,adadebay@umd.edu,false,2023-03-06 13:13,4.4 secs,Mapbox Satellite,,POINT(39.4560023573566 -3.758948134930454),357,Non-crop +358,358,34.19309376608378,-0.7112667390382308,adadebay@umd.edu,false,2023-03-06 13:50,18.0 secs,Mapbox Satellite,,POINT(34.19309376608378 -0.711266739038231),358,Crop +359,359,38.08520562034121,-0.7104578632257538,adadebay@umd.edu,false,2023-03-06 13:14,28.7 secs,Planet Monthly Mosaics,,POINT(38.08520562034121 -0.710457863225754),359,Non-crop +360,360,39.8764623690857,-0.41394507470955105,adadebay@umd.edu,false,2023-03-06 13:14,9.7 secs,Planet Monthly Mosaics,,POINT(39.8764623690857 -0.413945074709551),360,Non-crop +361,361,36.59090129950348,4.401266892103294,adadebay@umd.edu,false,2023-03-06 13:14,8.2 secs,Planet Monthly Mosaics,,POINT(36.59090129950348 4.401266892103294),361,Non-crop +362,362,36.80582306681952,3.5421413369810377,adadebay@umd.edu,false,2023-03-06 13:15,10.6 secs,Planet Monthly Mosaics,,POINT(36.80582306681952 3.542141336981038),362,Non-crop +363,363,40.84664060103932,1.7168775775104577,adadebay@umd.edu,false,2023-03-06 13:15,7.0 secs,Planet Monthly Mosaics,,POINT(40.84664060103932 1.716877577510458),363,Non-crop +364,364,40.68156500663407,1.129225977950579,adadebay@umd.edu,false,2023-03-06 13:15,6.1 secs,Planet Monthly Mosaics,,POINT(40.68156500663407 1.129225977950579),364,Non-crop +365,365,38.55642953523764,2.918040794846058,adadebay@umd.edu,false,2023-03-06 13:50,5.4 secs,Mapbox Satellite,,POINT(38.55642953523764 2.918040794846058),365,Non-crop +366,366,39.0709024919474,-3.7913328870442045,adadebay@umd.edu,false,2023-03-06 13:15,8.0 secs,Mapbox Satellite,,POINT(39.0709024919474 -3.791332887044204),366,Non-crop +367,367,36.638130724991306,-0.49460489559864695,adadebay@umd.edu,false,2023-03-06 13:15,4.6 secs,Mapbox Satellite,,POINT(36.638130724991306 -0.494604895598647),367,Non-crop +368,368,37.63079766174867,-0.053604990844186254,adadebay@umd.edu,false,2023-03-06 13:15,8.0 secs,Mapbox Satellite,,POINT(37.63079766174867 -0.053604990844186),368,Crop +369,369,35.827907745167956,-0.853811982262311,adadebay@umd.edu,false,2023-03-06 13:15,4.7 secs,Mapbox Satellite,,POINT(35.827907745167956 -0.853811982262311),369,Non-crop +370,370,39.167720267265224,1.9374319528339419,adadebay@umd.edu,false,2023-03-06 13:16,4.0 secs,Mapbox Satellite,,POINT(39.167720267265224 1.937431952833942),370,Non-crop +371,371,34.212762133967715,-0.8357804134194107,adadebay@umd.edu,false,2023-03-06 13:16,5.2 secs,Mapbox Satellite,,POINT(34.212762133967715 -0.835780413419411),371,Crop +372,372,37.96110516185624,0.14781455235668403,adadebay@umd.edu,false,2023-03-06 13:16,5.6 secs,Mapbox Satellite,,POINT(37.96110516185624 0.147814552356684),372,Non-crop +373,373,34.47463753982868,0.10990526138305981,adadebay@umd.edu,false,2023-03-06 13:17,80.6 secs,Sentinel-2,,POINT(34.47463753982868 0.10990526138306),373,Crop +374,374,39.69251931795742,-2.5541267260917877,adadebay@umd.edu,false,2023-03-06 13:17,14.6 secs,Mapbox Satellite,,POINT(39.69251931795742 -2.554126726091788),374,Non-crop +375,375,36.68244024793016,-1.8676442073681212,adadebay@umd.edu,false,2023-03-06 13:17,6.4 secs,Mapbox Satellite,,POINT(36.68244024793016 -1.867644207368121),375,Non-crop +376,376,36.853159792048174,3.4206556695885375,adadebay@umd.edu,false,2023-03-06 13:18,6.4 secs,Mapbox Satellite,,POINT(36.853159792048174 3.420655669588538),376,Non-crop +377,377,38.92586119879842,1.3194550813766501,adadebay@umd.edu,false,2023-03-06 13:18,4.8 secs,Mapbox Satellite,,POINT(38.92586119879842 1.31945508137665),377,Non-crop +378,378,38.69999902903868,2.7545836053240813,adadebay@umd.edu,false,2023-03-06 13:18,20.7 secs,Mapbox Satellite,,POINT(38.69999902903868 2.754583605324081),378,Non-crop +379,379,34.83181317134586,1.144867592720978,adadebay@umd.edu,false,2023-03-06 13:18,5.1 secs,Mapbox Satellite,,POINT(34.83181317134586 1.144867592720978),379,Crop +380,380,40.90693313748294,1.8439313546702625,adadebay@umd.edu,false,2023-03-06 13:18,4.5 secs,Mapbox Satellite,,POINT(40.90693313748294 1.843931354670262),380,Non-crop +381,381,34.1304218109236,-0.5631023328537231,adadebay@umd.edu,false,2023-03-06 13:18,7.5 secs,Mapbox Satellite,,POINT(34.1304218109236 -0.563102332853723),381,Non-crop +382,382,39.200957664948234,-2.5959413740347235,adadebay@umd.edu,false,2023-03-06 13:18,6.1 secs,Mapbox Satellite,,POINT(39.200957664948234 -2.595941374034724),382,Non-crop +383,383,38.63760081316255,1.1224819906657701,adadebay@umd.edu,false,2023-03-06 13:18,5.3 secs,Mapbox Satellite,,POINT(38.63760081316255 1.12248199066577),383,Non-crop +384,384,34.795026265560566,0.6743079317838352,adadebay@umd.edu,false,2023-03-06 13:19,6.2 secs,Mapbox Satellite,,POINT(34.795026265560566 0.674307931783835),384,Crop +385,385,40.272953061237764,-1.7216502223117192,adadebay@umd.edu,false,2023-03-06 13:19,5.1 secs,Mapbox Satellite,,POINT(40.272953061237764 -1.721650222311719),385,Non-crop +386,386,34.511216655298945,-0.40840217196105755,adadebay@umd.edu,false,2023-03-06 13:19,6.7 secs,Mapbox Satellite,,POINT(34.511216655298945 -0.408402171961058),386,Non-crop +387,387,38.532332486665574,-3.4722325425673475,adadebay@umd.edu,false,2023-03-06 13:19,5.4 secs,Mapbox Satellite,,POINT(38.532332486665574 -3.472232542567348),387,Non-crop +388,388,34.824113704860466,1.2169539748385623,adadebay@umd.edu,false,2023-03-06 13:19,4.5 secs,Mapbox Satellite,,POINT(34.824113704860466 1.216953974838562),388,Crop +389,389,37.388102475893305,-0.7540557617057277,adadebay@umd.edu,false,2023-03-06 13:19,8.1 secs,Mapbox Satellite,,POINT(37.388102475893305 -0.754055761705728),389,Crop +390,390,37.42891460399352,-0.45483461285550036,adadebay@umd.edu,false,2023-03-06 13:19,8.2 secs,Mapbox Satellite,,POINT(37.42891460399352 -0.4548346128555),390,Non-crop +391,391,36.967887534726245,-0.8095488907087274,adadebay@umd.edu,false,2023-03-06 13:19,5.2 secs,Mapbox Satellite,,POINT(36.967887534726245 -0.809548890708727),391,Non-crop +392,392,35.527048399962815,4.0989025202231,adadebay@umd.edu,false,2023-03-06 13:19,6.2 secs,Mapbox Satellite,,POINT(35.527048399962815 4.0989025202231),392,Non-crop +393,393,36.924922755405284,-1.1232711340297845,adadebay@umd.edu,false,2023-03-06 13:20,11.5 secs,Mapbox Satellite,,POINT(36.924922755405284 -1.123271134029784),393,Non-crop +394,394,37.257050079664545,0.7989869091198432,adadebay@umd.edu,false,2023-03-06 13:20,5.1 secs,Mapbox Satellite,,POINT(37.257050079664545 0.798986909119843),394,Non-crop +395,395,38.95454874266865,3.252171003985754,adadebay@umd.edu,false,2023-03-06 13:20,5.2 secs,Mapbox Satellite,,POINT(38.95454874266865 3.252171003985754),395,Non-crop +396,396,35.341565025437944,-0.29359632555640297,adadebay@umd.edu,false,2023-03-06 13:20,28.5 secs,Mapbox Satellite,,POINT(35.341565025437944 -0.293596325556403),396,Non-crop +397,397,35.79074795959633,-1.9464192254714852,adadebay@umd.edu,false,2023-03-06 13:20,6.7 secs,Mapbox Satellite,,POINT(35.79074795959633 -1.946419225471485),397,Non-crop +398,398,34.651917881792194,0.5319953940694888,adadebay@umd.edu,false,2023-03-06 13:20,5.1 secs,Mapbox Satellite,,POINT(34.651917881792194 0.531995394069489),398,Non-crop +399,399,41.1306620114978,3.1745220367098654,adadebay@umd.edu,false,2023-03-06 13:21,5.2 secs,Mapbox Satellite,,POINT(41.1306620114978 3.174522036709865),399,Non-crop +400,400,34.4371951436428,-0.7709979914980767,adadebay@umd.edu,false,2023-03-06 13:21,4.8 secs,Mapbox Satellite,,POINT(34.4371951436428 -0.770997991498077),400,Non-crop +401,401,34.558043775685306,3.125264995688962,adadebay@umd.edu,false,2023-03-06 13:21,5.1 secs,Mapbox Satellite,,POINT(34.558043775685306 3.125264995688962),401,Non-crop +402,402,37.981040475396064,3.6028567914524987,adadebay@umd.edu,false,2023-03-06 13:21,5.6 secs,Mapbox Satellite,,POINT(37.981040475396064 3.602856791452499),402,Non-crop +403,403,38.14269055562707,0.6772356097837042,adadebay@umd.edu,false,2023-03-06 13:21,4.3 secs,Mapbox Satellite,,POINT(38.14269055562707 0.677235609783704),403,Non-crop +404,404,34.46795120792515,0.6249513233906141,adadebay@umd.edu,false,2023-03-06 13:21,9.2 secs,Mapbox Satellite,,POINT(34.46795120792515 0.624951323390614),404,Crop +405,405,35.18905016180328,-0.9303939864286976,adadebay@umd.edu,false,2023-03-06 13:21,4.5 secs,Mapbox Satellite,,POINT(35.18905016180328 -0.930393986428698),405,Crop +406,406,34.680031744225815,-1.056354575251415,adadebay@umd.edu,false,2023-03-06 13:21,10.6 secs,Mapbox Satellite,,POINT(34.680031744225815 -1.056354575251415),406,Non-crop +407,407,38.584107662616205,-4.000071152288873,adadebay@umd.edu,false,2023-03-06 13:21,5.5 secs,Mapbox Satellite,,POINT(38.584107662616205 -4.000071152288873),407,Non-crop +408,408,34.54922635792607,0.568676232467724,adadebay@umd.edu,false,2023-03-06 13:22,6.8 secs,Mapbox Satellite,,POINT(34.54922635792607 0.568676232467724),408,Non-crop +409,409,34.46976620330974,0.5891906542491239,adadebay@umd.edu,false,2023-03-06 13:22,5.0 secs,Mapbox Satellite,,POINT(34.46976620330974 0.589190654249124),409,Crop +410,410,34.317026783547334,-0.14004161925952469,adadebay@umd.edu,false,2023-03-06 13:22,6.1 secs,Mapbox Satellite,,POINT(34.317026783547334 -0.140041619259525),410,Crop +411,411,37.67958111982131,-0.031077482460660336,adadebay@umd.edu,false,2023-03-06 13:22,5.2 secs,Mapbox Satellite,,POINT(37.67958111982131 -0.03107748246066),411,Non-crop +412,412,34.97698104242654,0.5048837047733515,adadebay@umd.edu,false,2023-03-06 13:23,76.0 secs,Mapbox Satellite,,POINT(34.97698104242654 0.504883704773352),412,Non-crop +413,413,39.42808479465304,2.647284919989776,adadebay@umd.edu,false,2023-03-06 13:23,6.3 secs,Mapbox Satellite,,POINT(39.42808479465304 2.647284919989776),413,Non-crop +414,414,36.941839061604995,3.049551867540078,adadebay@umd.edu,false,2023-03-06 13:23,5.4 secs,Mapbox Satellite,,POINT(36.941839061604995 3.049551867540078),414,Non-crop +415,415,34.79602405819177,0.09970469632021167,adadebay@umd.edu,false,2023-03-06 13:23,6.1 secs,Mapbox Satellite,,POINT(34.79602405819177 0.099704696320212),415,Non-crop +416,416,34.566268622694864,-0.9309187229149422,adadebay@umd.edu,false,2023-03-06 13:24,23.4 secs,Sentinel-2,,POINT(34.566268622694864 -0.930918722914942),416,Crop +417,417,34.193493927332696,-0.7542905932579258,adadebay@umd.edu,false,2023-03-06 13:24,30.9 secs,Mapbox Satellite,,POINT(34.193493927332696 -0.754290593257926),417,Non-crop +418,418,37.15096183503388,-0.33350560480048325,adadebay@umd.edu,false,2023-03-06 13:24,5.7 secs,Mapbox Satellite,,POINT(37.15096183503388 -0.333505604800483),418,Crop +419,419,34.51308490339396,-0.7547247822555792,adadebay@umd.edu,false,2023-03-06 13:25,6.8 secs,Mapbox Satellite,,POINT(34.51308490339396 -0.754724782255579),419,Crop +420,420,36.50791814531247,-0.7870475476290038,adadebay@umd.edu,false,2023-03-06 13:26,116.6 secs,Sentinel-2,,POINT(36.50791814531247 -0.787047547629004),420,Crop +421,421,37.82868384399238,3.714236460960642,adadebay@umd.edu,false,2023-03-06 13:27,8.4 secs,Mapbox Satellite,,POINT(37.82868384399238 3.714236460960642),421,Non-crop +422,422,40.009040766686724,-0.3419198637588498,adadebay@umd.edu,false,2023-03-06 13:27,6.2 secs,Mapbox Satellite,,POINT(40.009040766686724 -0.34191986375885),422,Non-crop +423,423,34.491276416658955,-1.2346310142928727,adadebay@umd.edu,false,2023-03-06 13:27,5.6 secs,Mapbox Satellite,,POINT(34.491276416658955 -1.234631014292873),423,Crop +424,424,35.15441964409439,-0.43733764965412875,adadebay@umd.edu,false,2023-03-06 13:27,5.3 secs,Mapbox Satellite,,POINT(35.15441964409439 -0.437337649654129),424,Non-crop +425,425,35.28686492785766,4.213669588373389,adadebay@umd.edu,false,2023-03-06 13:27,7.9 secs,Mapbox Satellite,,POINT(35.28686492785766 4.213669588373389),425,Non-crop +426,426,40.499705577963645,3.056516667636345,adadebay@umd.edu,false,2023-03-06 13:27,4.4 secs,Mapbox Satellite,,POINT(40.499705577963645 3.056516667636345),426,Non-crop +427,427,40.30464352437439,-0.4326296899359431,adadebay@umd.edu,false,2023-03-06 13:27,8.6 secs,Mapbox Satellite,,POINT(40.30464352437439 -0.432629689935943),427,Non-crop +428,428,39.34917441922158,1.099228413243411,adadebay@umd.edu,false,2023-03-06 13:27,13.8 secs,Mapbox Satellite,,POINT(39.34917441922158 1.099228413243411),428,Non-crop +429,429,34.26078466118539,-1.0247476972595801,adadebay@umd.edu,false,2023-03-06 13:28,38.0 secs,Mapbox Satellite,,POINT(34.26078466118539 -1.02474769725958),429,Crop +430,430,39.74078659873234,-1.1870999208832935,adadebay@umd.edu,false,2023-03-06 13:28,7.2 secs,Mapbox Satellite,,POINT(39.74078659873234 -1.187099920883294),430,Non-crop +431,431,34.76303145495171,0.4669028149604851,adadebay@umd.edu,false,2023-03-06 13:28,5.2 secs,Mapbox Satellite,,POINT(34.76303145495171 0.466902814960485),431,Crop +432,432,37.39470110843973,-0.5501557583050474,adadebay@umd.edu,false,2023-03-06 13:28,4.8 secs,Mapbox Satellite,,POINT(37.39470110843973 -0.550155758305047),432,Non-crop +433,433,37.174734166271435,-2.02859100605471,adadebay@umd.edu,false,2023-03-06 13:28,4.7 secs,Mapbox Satellite,,POINT(37.174734166271435 -2.02859100605471),433,Non-crop +434,434,35.7713835612932,1.7742478484709758,adadebay@umd.edu,false,2023-03-06 13:29,5.1 secs,Mapbox Satellite,,POINT(35.7713835612932 1.774247848470976),434,Non-crop +435,435,35.054458324171065,-0.44070243296815076,adadebay@umd.edu,false,2023-03-06 13:29,7.3 secs,Mapbox Satellite,,POINT(35.054458324171065 -0.440702432968151),435,Crop +436,436,35.535317107443134,4.0300216964176565,adadebay@umd.edu,false,2023-03-06 13:29,4.8 secs,Mapbox Satellite,,POINT(35.535317107443134 4.030021696417656),436,Non-crop +437,437,36.195615641572736,1.2500737632246104,adadebay@umd.edu,false,2023-03-06 13:29,8.0 secs,Mapbox Satellite,,POINT(36.195615641572736 1.25007376322461),437,Non-crop +438,438,34.585500564173465,-1.1006159436375704,adadebay@umd.edu,false,2023-03-06 13:29,5.2 secs,Mapbox Satellite,,POINT(34.585500564173465 -1.10061594363757),438,Crop +439,439,36.93681116460859,-0.8568247999425626,adadebay@umd.edu,false,2023-03-06 13:29,10.5 secs,Mapbox Satellite,,POINT(36.93681116460859 -0.856824799942563),439,Non-crop +440,440,34.653940619977,-1.2979980707441048,adadebay@umd.edu,false,2023-03-06 13:29,6.5 secs,Mapbox Satellite,,POINT(34.653940619977 -1.297998070744105),440,Non-crop +441,441,39.95325288909631,-3.0539655940358257,adadebay@umd.edu,false,2023-03-06 13:29,5.1 secs,Mapbox Satellite,,POINT(39.95325288909631 -3.053965594035826),441,Non-crop +442,442,34.79643269660636,0.954060045767657,adadebay@umd.edu,false,2023-03-06 13:30,12.3 secs,Mapbox Satellite,,POINT(34.79643269660636 0.954060045767657),442,Non-crop +443,443,37.64560569646061,-0.4995748989339322,adadebay@umd.edu,false,2023-03-06 13:30,14.1 secs,Mapbox Satellite,,POINT(37.64560569646061 -0.499574898933932),443,Crop +444,444,35.193697555577735,-1.0090949072315327,adadebay@umd.edu,false,2023-03-06 13:30,7.9 secs,Mapbox Satellite,,POINT(35.193697555577735 -1.009094907231533),444,Non-crop +445,445,34.29444546033131,-0.1648308472018498,adadebay@umd.edu,false,2023-03-06 13:30,10.6 secs,Mapbox Satellite,,POINT(34.29444546033131 -0.16483084720185),445,Crop +446,446,34.314974037204585,-0.8050913043294229,adadebay@umd.edu,false,2023-03-06 13:30,7.1 secs,Mapbox Satellite,,POINT(34.314974037204585 -0.805091304329423),446,Non-crop +447,447,40.31294294731426,-1.7926326213420247,adadebay@umd.edu,false,2023-03-06 13:30,4.8 secs,Mapbox Satellite,,POINT(40.31294294731426 -1.792632621342025),447,Non-crop +448,448,36.91616303309522,-2.5727932209337974,adadebay@umd.edu,false,2023-03-06 13:30,6.1 secs,Mapbox Satellite,,POINT(36.91616303309522 -2.572793220933797),448,Non-crop +449,449,40.3485941388307,-1.8043211728138038,adadebay@umd.edu,false,2023-03-06 13:31,4.6 secs,Mapbox Satellite,,POINT(40.3485941388307 -1.804321172813804),449,Non-crop +450,450,34.75482976822808,1.2027323213729808,adadebay@umd.edu,false,2023-03-06 13:31,5.2 secs,Mapbox Satellite,,POINT(34.75482976822808 1.202732321372981),450,Non-crop +451,451,35.039639014761306,4.2483244000980145,adadebay@umd.edu,false,2023-03-06 13:31,5.1 secs,Mapbox Satellite,,POINT(35.039639014761306 4.248324400098014),451,Non-crop +452,452,40.2874788432159,2.504938312987063,adadebay@umd.edu,false,2023-03-06 13:31,4.8 secs,Mapbox Satellite,,POINT(40.2874788432159 2.504938312987063),452,Non-crop +453,453,37.573692021019895,-0.004673769817114266,adadebay@umd.edu,false,2023-03-06 13:31,5.9 secs,Mapbox Satellite,,POINT(37.573692021019895 -0.004673769817114),453,Crop +454,454,36.807037684141676,2.542819879774906,adadebay@umd.edu,false,2023-03-06 13:31,5.3 secs,Mapbox Satellite,,POINT(36.807037684141676 2.542819879774906),454,Non-crop +455,455,37.0048015872321,-0.45869277751116916,adadebay@umd.edu,false,2023-03-06 13:31,8.2 secs,Mapbox Satellite,,POINT(37.0048015872321 -0.458692777511169),455,Non-crop +456,456,36.84196187497421,3.7249805816569435,adadebay@umd.edu,false,2023-03-06 13:31,4.8 secs,Mapbox Satellite,,POINT(36.84196187497421 3.724980581656944),456,Non-crop +457,457,34.649646702806294,0.6912557299441862,adadebay@umd.edu,false,2023-03-06 13:31,8.1 secs,Mapbox Satellite,,POINT(34.649646702806294 0.691255729944186),457,Crop +458,458,39.27351047995854,2.382870821022328,adadebay@umd.edu,false,2023-03-06 13:31,4.9 secs,Mapbox Satellite,,POINT(39.27351047995854 2.382870821022328),458,Non-crop +459,459,34.09255046019023,0.13750458019816195,adadebay@umd.edu,false,2023-03-06 13:32,6.0 secs,Mapbox Satellite,,POINT(34.09255046019023 0.137504580198162),459,Non-crop +460,460,35.196367961673985,-0.9384964043103956,adadebay@umd.edu,false,2023-03-06 13:32,5.4 secs,Mapbox Satellite,,POINT(35.196367961673985 -0.938496404310396),460,Non-crop +461,461,34.729242476157864,-0.006280115060495967,adadebay@umd.edu,false,2023-03-06 13:32,15.0 secs,Mapbox Satellite,,POINT(34.729242476157864 -0.006280115060496),461,Non-crop +462,462,37.14320247144663,-0.5799924116994661,adadebay@umd.edu,false,2023-03-06 13:32,5.6 secs,Mapbox Satellite,,POINT(37.14320247144663 -0.579992411699466),462,Non-crop +463,463,34.839386313724425,-0.5907161983375747,adadebay@umd.edu,false,2023-03-06 13:32,6.4 secs,Mapbox Satellite,,POINT(34.839386313724425 -0.590716198337575),463,Crop +464,464,37.298134606247274,-2.7530158736682773,adadebay@umd.edu,false,2023-03-06 13:32,5.6 secs,Mapbox Satellite,,POINT(37.298134606247274 -2.753015873668277),464,Non-crop +465,465,37.235080182010115,3.2939567840656823,adadebay@umd.edu,false,2023-03-06 13:32,6.4 secs,Mapbox Satellite,,POINT(37.235080182010115 3.293956784065682),465,Non-crop +466,466,34.620563644213945,3.1399987547652173,adadebay@umd.edu,false,2023-03-06 13:32,5.0 secs,Mapbox Satellite,,POINT(34.620563644213945 3.139998754765217),466,Non-crop +467,467,37.90330112445743,2.0453183232955094,adadebay@umd.edu,false,2023-03-06 13:32,5.0 secs,Mapbox Satellite,,POINT(37.90330112445743 2.045318323295509),467,Non-crop +468,468,38.75219866912606,1.4087696247600232,adadebay@umd.edu,false,2023-03-06 13:33,5.0 secs,Mapbox Satellite,,POINT(38.75219866912606 1.408769624760023),468,Non-crop +469,469,36.09478312311987,-0.45270767372427245,adadebay@umd.edu,false,2023-03-06 13:33,10.5 secs,Mapbox Satellite,,POINT(36.09478312311987 -0.452707673724272),469,Non-crop +470,470,39.14544637042974,3.0526918285416156,adadebay@umd.edu,false,2023-03-06 13:33,5.7 secs,Mapbox Satellite,,POINT(39.14544637042974 3.052691828541616),470,Non-crop +471,471,36.986050045933645,0.8166083567094135,adadebay@umd.edu,false,2023-03-06 13:33,26.4 secs,Mapbox Satellite,,POINT(36.986050045933645 0.816608356709414),471,Non-crop +472,472,36.67508229636412,2.360931840234958,adadebay@umd.edu,false,2023-03-06 13:33,4.1 secs,Mapbox Satellite,,POINT(36.67508229636412 2.360931840234958),472,Non-crop +473,473,34.87929322199816,-1.3410826721831222,adadebay@umd.edu,false,2023-03-06 13:33,6.7 secs,Mapbox Satellite,,POINT(34.87929322199816 -1.341082672183122),473,Non-crop +474,474,36.96034269804192,0.6752585022986302,adadebay@umd.edu,false,2023-03-06 13:34,5.8 secs,Mapbox Satellite,,POINT(36.96034269804192 0.67525850229863),474,Non-crop +475,475,38.01094738788046,0.4034110699662577,adadebay@umd.edu,false,2023-03-06 13:34,5.2 secs,Mapbox Satellite,,POINT(38.01094738788046 0.403411069966258),475,Non-crop +476,476,37.678195029831784,0.4669841625774332,adadebay@umd.edu,false,2023-03-06 13:34,5.2 secs,Mapbox Satellite,,POINT(37.678195029831784 0.466984162577433),476,Non-crop +477,477,35.56489986263098,4.17275389765291,adadebay@umd.edu,false,2023-03-06 13:34,4.2 secs,Mapbox Satellite,,POINT(35.56489986263098 4.17275389765291),477,Non-crop +478,478,38.89991542855711,-0.10599067675841689,adadebay@umd.edu,false,2023-03-06 13:34,4.8 secs,Mapbox Satellite,,POINT(38.89991542855711 -0.105990676758417),478,Non-crop +479,479,38.16088744929678,-0.5340005076082127,adadebay@umd.edu,false,2023-03-06 13:34,26.5 secs,Sentinel-2,,POINT(38.16088744929678 -0.534000507608213),479,Non-crop +480,480,36.31334231145933,1.7564825870660954,adadebay@umd.edu,false,2023-03-06 13:35,10.2 secs,Mapbox Satellite,,POINT(36.31334231145933 1.756482587066095),480,Non-crop +481,481,36.05446021526722,-0.7701864509339067,adadebay@umd.edu,false,2023-03-06 13:35,18.1 secs,Mapbox Satellite,,POINT(36.05446021526722 -0.770186450933907),481,Non-crop +482,482,34.634601312187485,0.08943773979142051,adadebay@umd.edu,false,2023-03-06 13:35,6.4 secs,Mapbox Satellite,,POINT(34.634601312187485 0.089437739791421),482,Crop +483,483,37.91044452150111,3.5311304324163824,adadebay@umd.edu,false,2023-03-06 13:35,5.0 secs,Mapbox Satellite,,POINT(37.91044452150111 3.531130432416382),483,Non-crop +484,484,34.5575298635729,-1.2831810660561516,adadebay@umd.edu,false,2023-03-06 13:35,12.7 secs,Mapbox Satellite,,POINT(34.5575298635729 -1.283181066056152),484,Crop +485,485,35.39516029245672,-0.8794258046257776,adadebay@umd.edu,false,2023-03-06 13:36,22.2 secs,Mapbox Satellite,,POINT(35.39516029245672 -0.879425804625778),485,Non-crop +486,486,38.321928188337154,3.043383821526915,adadebay@umd.edu,false,2023-03-06 13:36,5.4 secs,Mapbox Satellite,,POINT(38.321928188337154 3.043383821526915),486,Non-crop +487,487,39.34805366474005,0.5760758805466408,adadebay@umd.edu,false,2023-03-06 13:36,4.7 secs,Mapbox Satellite,,POINT(39.34805366474005 0.576075880546641),487,Non-crop +488,488,40.70228392102145,2.2930049702457507,adadebay@umd.edu,false,2023-03-06 13:36,4.1 secs,Mapbox Satellite,,POINT(40.70228392102145 2.293004970245751),488,Non-crop +489,489,34.988652426404,2.685077579713715,adadebay@umd.edu,false,2023-03-06 13:36,5.0 secs,Mapbox Satellite,,POINT(34.988652426404 2.685077579713715),489,Non-crop +490,490,35.0036305876314,-0.7040441877978902,adadebay@umd.edu,false,2023-03-06 13:36,9.7 secs,Mapbox Satellite,,POINT(35.0036305876314 -0.70404418779789),490,Non-crop +491,491,34.653443139440675,-0.02515374865425375,adadebay@umd.edu,false,2023-03-06 13:36,5.4 secs,Mapbox Satellite,,POINT(34.653443139440675 -0.025153748654254),491,Non-crop +492,492,40.614414543790275,0.3435724730712614,adadebay@umd.edu,false,2023-03-06 13:36,6.3 secs,Mapbox Satellite,,POINT(40.614414543790275 0.343572473071261),492,Non-crop +493,493,38.51702418202515,-2.8238115957339627,adadebay@umd.edu,false,2023-03-06 13:36,4.8 secs,Mapbox Satellite,,POINT(38.51702418202515 -2.823811595733963),493,Non-crop +494,494,34.383934200549206,4.452384575378512,adadebay@umd.edu,false,2023-03-06 13:36,4.5 secs,Mapbox Satellite,,POINT(34.383934200549206 4.452384575378512),494,Non-crop +495,495,37.70530305115963,2.5868444101204076,adadebay@umd.edu,false,2023-03-06 13:37,3.7 secs,Mapbox Satellite,,POINT(37.70530305115963 2.586844410120408),495,Non-crop +496,496,35.33655489953117,-1.367002453719124,adadebay@umd.edu,false,2023-03-06 13:37,4.0 secs,Mapbox Satellite,,POINT(35.33655489953117 -1.367002453719124),496,Non-crop +497,497,38.39506049495438,2.5714102466385724,adadebay@umd.edu,false,2023-03-06 13:37,4.7 secs,Mapbox Satellite,,POINT(38.39506049495438 2.571410246638572),497,Non-crop +498,498,39.084849466631404,3.2272009282858454,adadebay@umd.edu,false,2023-03-06 13:37,5.1 secs,Mapbox Satellite,,POINT(39.084849466631404 3.227200928285845),498,Non-crop +499,499,37.09093189496214,0.9915204603295135,adadebay@umd.edu,false,2023-03-06 13:37,4.1 secs,Mapbox Satellite,,POINT(37.09093189496214 0.991520460329514),499,Non-crop +500,500,37.349874791030274,-2.6561425061602493,adadebay@umd.edu,false,2023-03-06 13:37,5.0 secs,Mapbox Satellite,,POINT(37.349874791030274 -2.656142506160249),500,Non-crop +501,501,35.17501834522299,-0.7515855198732806,adadebay@umd.edu,false,2023-03-06 13:38,74.6 secs,Mapbox Satellite,,POINT(35.17501834522299 -0.751585519873281),501,Non-crop +502,502,35.71291563450446,-0.6426003621868701,adadebay@umd.edu,false,2023-03-06 13:38,5.3 secs,Mapbox Satellite,,POINT(35.71291563450446 -0.64260036218687),502,Non-crop +503,503,34.73484872536813,-0.8482526044699542,adadebay@umd.edu,false,2023-03-06 13:38,6.2 secs,Mapbox Satellite,,POINT(34.73484872536813 -0.848252604469954),503,Crop +504,504,39.023752295479596,-1.6523111301971936,adadebay@umd.edu,false,2023-03-06 13:38,4.8 secs,Mapbox Satellite,,POINT(39.023752295479596 -1.652311130197194),504,Non-crop +505,505,40.78764234677859,2.693632786095179,adadebay@umd.edu,false,2023-03-06 13:38,4.4 secs,Mapbox Satellite,,POINT(40.78764234677859 2.693632786095179),505,Non-crop +506,506,35.06000147333601,-0.7966242423341563,adadebay@umd.edu,false,2023-03-06 13:39,5.3 secs,Mapbox Satellite,,POINT(35.06000147333601 -0.796624242334156),506,Non-crop +507,507,36.30009993341707,-0.8232024117545933,adadebay@umd.edu,false,2023-03-06 13:39,5.0 secs,Mapbox Satellite,,POINT(36.30009993341707 -0.823202411754593),507,Non-crop +508,508,35.10989219436885,-0.32787050503899345,adadebay@umd.edu,false,2023-03-06 13:39,21.2 secs,Mapbox Satellite,,POINT(35.10989219436885 -0.327870505038993),508,Crop +509,509,34.828144212308935,-0.5773129039263161,adadebay@umd.edu,false,2023-03-06 13:40,72.7 secs,Mapbox Satellite,,POINT(34.828144212308935 -0.577312903926316),509,Non-crop +510,510,35.97078220549668,2.9560655204795507,adadebay@umd.edu,false,2023-03-06 13:40,6.6 secs,Mapbox Satellite,,POINT(35.97078220549668 2.956065520479551),510,Non-crop +511,511,35.10284238847304,-0.4966636896018876,adadebay@umd.edu,false,2023-03-06 13:40,6.3 secs,Mapbox Satellite,,POINT(35.10284238847304 -0.496663689601888),511,Non-crop +512,512,40.132913213558695,-3.146251638192202,adadebay@umd.edu,false,2023-03-06 13:41,5.7 secs,Mapbox Satellite,,POINT(40.132913213558695 -3.146251638192202),512,Non-crop +513,513,34.22388893299226,0.605241533460933,adadebay@umd.edu,false,2023-03-06 13:41,6.7 secs,Mapbox Satellite,,POINT(34.22388893299226 0.605241533460933),513,Crop +514,514,38.96670779650444,-0.4182302700090625,adadebay@umd.edu,false,2023-03-06 13:41,4.4 secs,Mapbox Satellite,,POINT(38.96670779650444 -0.418230270009062),514,Non-crop +515,515,40.32965437189055,1.3051310381454755,adadebay@umd.edu,false,2023-03-06 13:41,4.0 secs,Mapbox Satellite,,POINT(40.32965437189055 1.305131038145476),515,Non-crop +516,516,34.96478463720174,-0.5794803589111308,adadebay@umd.edu,false,2023-03-06 13:41,8.0 secs,Mapbox Satellite,,POINT(34.96478463720174 -0.579480358911131),516,Crop +517,517,40.96226183600945,-0.2882607556531856,adadebay@umd.edu,false,2023-03-06 13:41,5.5 secs,Mapbox Satellite,,POINT(40.96226183600945 -0.288260755653186),517,Non-crop +518,518,39.735173373793074,3.116710754764235,adadebay@umd.edu,false,2023-03-06 13:41,6.6 secs,Mapbox Satellite,,POINT(39.735173373793074 3.116710754764235),518,Non-crop +519,519,36.82035819397363,1.9556658604114368,adadebay@umd.edu,false,2023-03-06 13:41,4.2 secs,Mapbox Satellite,,POINT(36.82035819397363 1.955665860411437),519,Non-crop +520,520,35.35383455962234,2.211003394302324,adadebay@umd.edu,false,2023-03-06 13:41,3.8 secs,Mapbox Satellite,,POINT(35.35383455962234 2.211003394302324),520,Non-crop +521,521,35.410735825610544,-1.428738022679518,adadebay@umd.edu,false,2023-03-06 13:41,4.9 secs,Mapbox Satellite,,POINT(35.410735825610544 -1.428738022679518),521,Non-crop +522,522,37.19939040219627,2.0807162473361056,adadebay@umd.edu,false,2023-03-06 13:41,5.9 secs,Mapbox Satellite,,POINT(37.19939040219627 2.080716247336106),522,Non-crop +523,523,35.08208831520865,-0.9479913632950039,adadebay@umd.edu,false,2023-03-06 13:43,78.6 secs,Mapbox Satellite,,POINT(35.08208831520865 -0.947991363295004),523,Crop +524,524,36.529750333470275,-1.2510182266385572,adadebay@umd.edu,false,2023-03-06 13:43,4.9 secs,Mapbox Satellite,,POINT(36.529750333470275 -1.251018226638557),524,Non-crop +525,525,34.93545090273626,1.3595428403458083,adadebay@umd.edu,false,2023-03-06 13:43,5.2 secs,Mapbox Satellite,,POINT(34.93545090273626 1.359542840345808),525,Non-crop +526,526,34.71477847358641,0.00774202647017627,adadebay@umd.edu,false,2023-03-06 13:44,56.7 secs,Sentinel-2,,POINT(34.71477847358641 0.007742026470176),526,Crop +527,527,39.59186188819468,-1.1566710064881998,adadebay@umd.edu,false,2023-03-06 13:44,5.9 secs,Mapbox Satellite,,POINT(39.59186188819468 -1.1566710064882),527,Non-crop +528,528,38.23861965981843,0.33277349267400996,adadebay@umd.edu,false,2023-03-06 13:44,5.1 secs,Mapbox Satellite,,POINT(38.23861965981843 0.33277349267401),528,Non-crop +529,529,37.01526073538915,-0.7368748435057784,adadebay@umd.edu,false,2023-03-06 13:44,5.9 secs,Mapbox Satellite,,POINT(37.01526073538915 -0.736874843505778),529,Non-crop +530,530,38.09145658119295,-2.415198363979528,adadebay@umd.edu,false,2023-03-06 13:45,32.7 secs,Planet Monthly Mosaics,,POINT(38.09145658119295 -2.415198363979528),530,Crop +531,531,37.65093779234663,-0.026120886979007708,adadebay@umd.edu,false,2023-03-06 13:45,11.2 secs,Mapbox Satellite,,POINT(37.65093779234663 -0.026120886979008),531,Non-crop +532,532,40.536095203749554,3.632313062717191,adadebay@umd.edu,false,2023-03-06 13:45,12.5 secs,Mapbox Satellite,,POINT(40.536095203749554 3.632313062717191),532,Non-crop +533,533,34.56842069799163,0.8483574983470237,adadebay@umd.edu,false,2023-03-06 13:46,13.4 secs,Mapbox Satellite,,POINT(34.56842069799163 0.848357498347024),533,Crop +534,534,34.560156122324,-0.01247945700671197,adadebay@umd.edu,false,2023-03-06 13:46,16.9 secs,Mapbox Satellite,,POINT(34.560156122324 -0.012479457006712),534,Crop +535,535,36.52662305326271,1.9856940678472765,adadebay@umd.edu,false,2023-03-06 13:46,4.9 secs,Mapbox Satellite,,POINT(36.52662305326271 1.985694067847276),535,Non-crop +536,536,36.814465210482616,-1.0359101389648762,adadebay@umd.edu,false,2023-03-06 13:46,13.4 secs,Mapbox Satellite,,POINT(36.814465210482616 -1.035910138964876),536,Crop +537,537,37.06018956428111,-0.4699689113578657,adadebay@umd.edu,false,2023-03-06 13:46,5.9 secs,Mapbox Satellite,,POINT(37.06018956428111 -0.469968911357866),537,Non-crop +538,538,39.14612330417667,2.7377571101253593,adadebay@umd.edu,false,2023-03-06 13:46,6.0 secs,Mapbox Satellite,,POINT(39.14612330417667 2.737757110125359),538,Non-crop +539,539,37.87082220460906,-1.7089443100333617,adadebay@umd.edu,false,2023-03-06 13:47,18.9 secs,Mapbox Satellite,,POINT(37.87082220460906 -1.708944310033362),539,Non-crop +540,540,34.88255928203005,-0.023363158733180508,adadebay@umd.edu,false,2023-03-06 13:47,5.9 secs,Mapbox Satellite,,POINT(34.88255928203005 -0.023363158733181),540,Non-crop +541,541,34.46410083114127,-0.8173018845840876,adadebay@umd.edu,false,2023-03-06 13:49,151.5 secs,Mapbox Satellite,,POINT(34.46410083114127 -0.817301884584088),541,Non-crop +542,542,34.93492776673225,2.8648969176507055,adadebay@umd.edu,false,2023-03-06 13:49,7.8 secs,Mapbox Satellite,,POINT(34.93492776673225 2.864896917650706),542,Non-crop +543,543,38.81338965106459,1.119704331352858,adadebay@umd.edu,false,2023-03-06 13:50,5.7 secs,Mapbox Satellite,,POINT(38.81338965106459 1.119704331352858),543,Non-crop \ No newline at end of file From d213c33fd6e01cca8d67f5442410ae3563b4f515 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Sat, 23 Mar 2024 21:59:28 +0000 Subject: [PATCH 10/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- ...ion-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv | 2 +- ...ion-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv index 6da8c5d8..4f53ec0e 100644 --- a/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv +++ b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv @@ -542,4 +542,4 @@ 540,540,34.88255928203005,-0.023363158733180508,isha9a@umd.edu,false,2023-03-08 22:08,23.5 secs,Mapbox Satellite,,POINT(34.88255928203005 -0.023363158733181),540,Non-crop 541,541,34.46410083114127,-0.8173018845840876,isha9a@umd.edu,false,2023-03-08 22:09,77.0 secs,Mapbox Satellite,,POINT(34.46410083114127 -0.817301884584088),541,Crop 542,542,34.93492776673225,2.8648969176507055,isha9a@umd.edu,false,2023-03-08 22:09,10.7 secs,Mapbox Satellite,,POINT(34.93492776673225 2.864896917650706),542,Non-crop -543,543,38.81338965106459,1.119704331352858,isha9a@umd.edu,false,2023-03-08 22:09,9.2 secs,Mapbox Satellite,,POINT(38.81338965106459 1.119704331352858),543,Non-crop \ No newline at end of file +543,543,38.81338965106459,1.119704331352858,isha9a@umd.edu,false,2023-03-08 22:09,9.2 secs,Mapbox Satellite,,POINT(38.81338965106459 1.119704331352858),543,Non-crop diff --git a/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv index 44018a52..9087c61d 100644 --- a/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv +++ b/data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv @@ -542,4 +542,4 @@ 540,540,34.88255928203005,-0.023363158733180508,adadebay@umd.edu,false,2023-03-06 13:47,5.9 secs,Mapbox Satellite,,POINT(34.88255928203005 -0.023363158733181),540,Non-crop 541,541,34.46410083114127,-0.8173018845840876,adadebay@umd.edu,false,2023-03-06 13:49,151.5 secs,Mapbox Satellite,,POINT(34.46410083114127 -0.817301884584088),541,Non-crop 542,542,34.93492776673225,2.8648969176507055,adadebay@umd.edu,false,2023-03-06 13:49,7.8 secs,Mapbox Satellite,,POINT(34.93492776673225 2.864896917650706),542,Non-crop -543,543,38.81338965106459,1.119704331352858,adadebay@umd.edu,false,2023-03-06 13:50,5.7 secs,Mapbox Satellite,,POINT(38.81338965106459 1.119704331352858),543,Non-crop \ No newline at end of file +543,543,38.81338965106459,1.119704331352858,adadebay@umd.edu,false,2023-03-06 13:50,5.7 secs,Mapbox Satellite,,POINT(38.81338965106459 1.119704331352858),543,Non-crop From 2c1c5320e078d6bac1bf49d609cccf0b6a633637 Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Sat, 23 Mar 2024 19:32:58 -0400 Subject: [PATCH 11/21] Add Senegal area estimate --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 3692 +++++++++++++++++ 1 file changed, 3692 insertions(+) create mode 100644 maps/Senegal_2022/Senegal_area_estimate.ipynb diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb new file mode 100644 index 00000000..2cbf7b85 --- /dev/null +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -0,0 +1,3692 @@ +{ + "cells": [ + { + "cell_type": "markdown", + "metadata": { + "id": "778ff440" + }, + "source": [ + "# Intercomparison\n", + "\n", + "**Author:**\n", + "\n", + "**Last updated:**\n", + "\n", + "**Description:** Runs intercomparison for [Country Year]\n", + "\n", + "## 1. Setup" + ], + "id": "778ff440" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "fb42d13c" + }, + "outputs": [], + "source": [ + "# !earthengine authenticate" + ], + "id": "fb42d13c" + }, + { + "cell_type": "code", + "execution_count": 1, + "metadata": { + "id": "hZ8qzSlB75kl", + "colab": { + "base_uri": "https://localhost:8080/" + }, + "outputId": "80951fe8-fd11-4fd3-ce7a-f92012496b2d" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Cloning into 'crop-mask'...\n", + "remote: Enumerating objects: 12074, done.\u001b[K\n", + "remote: Counting objects: 100% (1485/1485), done.\u001b[K\n", + "remote: Compressing objects: 100% (449/449), done.\u001b[K\n", + "remote: Total 12074 (delta 1102), reused 1232 (delta 1009), pack-reused 10589\u001b[K\n", + "Receiving objects: 100% (12074/12074), 125.43 MiB | 11.56 MiB/s, done.\n", + "Resolving deltas: 100% (7824/7824), done.\n", + "Updating files: 100% (208/208), done.\n" + ] + } + ], + "source": [ + "!git clone https://github.com/nasaharvest/crop-mask.git" + ], + "id": "hZ8qzSlB75kl" + }, + { + "cell_type": "code", + "execution_count": 2, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "1fe-6D3f8LTb", + "outputId": "6c6848be-2e5f-4c10-ce9c-b4dc071a2795" + }, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "/content/crop-mask\n" + ] + } + ], + "source": [ + "%cd crop-mask/" + ], + "id": "1fe-6D3f8LTb" + }, + { + "cell_type": "code", + "source": [ + "!git checkout area-estimate-from-multi-land-cover" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/" + }, + "id": "V6lTs8Z9Pt-T", + "outputId": "a9ed0471-9de0-4299-b537-069aa07a453c" + }, + "id": "V6lTs8Z9Pt-T", + "execution_count": 3, + "outputs": [ + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Branch 'area-estimate-from-multi-land-cover' set up to track remote branch 'area-estimate-from-multi-land-cover' from 'origin'.\n", + "Switched to a new branch 'area-estimate-from-multi-land-cover'\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "gEUyxHk9MEU2" + }, + "outputs": [], + "source": [ + "!pip install cartopy -qq\n", + "!pip install rasterio -qq\n", + "!pip install dvc[gs] -qq" + ], + "id": "gEUyxHk9MEU2" + }, + { + "cell_type": "code", + "execution_count": 71, + "metadata": { + "id": "9907f9a5", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "outputId": "4e49e10a-1dc2-44f2-f39f-0acb74d3845b" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "import ee\n", + "import geemap\n", + "import sys\n", + "import pandas as pd\n", + "import numpy as np\n", + "\n", + "import geopandas as gpd\n", + "from pathlib import Path\n", + "\n", + "ee.Authenticate()\n", + "ee.Initialize(project=\"bsos-geog-harvest1\")\n", + "\n", + "sys.path.append(\"../..\")\n", + "\n", + "from src.compare_covermaps import TARGETS, filter_by_bounds, generate_report, CLASS_COL, COUNTRY_COL, get_ensemble_area\n", + "from src.compare_covermaps import TEST_COUNTRIES, TEST_CODE" + ], + "id": "9907f9a5" + }, + { + "cell_type": "markdown", + "metadata": { + "id": "c61ea4f8" + }, + "source": [ + "## 2. Read in evaluation set" + ], + "id": "c61ea4f8" + }, + { + "cell_type": "code", + "execution_count": 48, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "7f75e567", + "outputId": "ccea9d96-7484-4157-9cee-2058c97bbc02" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "country = \"Senegal\"\n", + "\n", + "if country not in TEST_CODE:\n", + " print(f\"WARNING: {country} not found in TEST_CODE in src/compare_covermaps.py\")\n", + "if country not in TEST_COUNTRIES:\n", + " print(f\"WARNING: {country} not found in TEST_COUNTRIES in src/compare_covermaps.py\")\n", + "if country not in TEST_CODE or country not in TEST_COUNTRIES:\n", + " print(\"Please update src/compare_covermaps.py and restart the notebook.\")\n", + "else:\n", + " country_code = TEST_CODE[country]\n", + " # dataset_path = \"../\" + TEST_COUNTRIES[country]" + ], + "id": "7f75e567" + }, + { + "cell_type": "code", + "execution_count": 35, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "prvHkUXTOe7P", + "outputId": "fe22a240-697a-4986-bfa7-f049935bd1cc" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "# dataset_path = TEST_COUNTRIES[country]\n", + "dataset_path = 'data/datasets/Senegal_CEO_2022.csv'" + ], + "id": "prvHkUXTOe7P" + }, + { + "cell_type": "code", + "source": [ + "# ceo_set1 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv'\n", + "# ceo_set2 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv'" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "66-YJBNxYAdF", + "outputId": "221a20ee-9808-4181-8354-d1877c544aca" + }, + "id": "66-YJBNxYAdF", + "execution_count": 7, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "def reference_sample_agree(ceo_ref1, ceo_ref2):\n", + " ceo_ref1 = pd.read_csv(ceo_ref1)\n", + " ceo_ref2 = pd.read_csv(ceo_ref2)\n", + "\n", + " assert ceo_ref1.columns[-1] == ceo_ref2.columns[-1]\n", + "\n", + " label_question = ceo_ref1.columns[-1]\n", + "\n", + " print(f\"Number of NANs/ missing answers in set 1: {ceo_ref1[label_question].isna().sum()}\")\n", + " print(f\"Number of NANs/ missing answers in set 2: {ceo_ref2[label_question].isna().sum()}\")\n", + "\n", + " if ceo_ref1.shape[0] != ceo_ref2.shape[0]:\n", + " print(\"The number of rows in the reference sets are not equal.\")\n", + " print(\"Checking for duplictes on 'plotid'..\")\n", + " print(\n", + " \" Number of duplicated in set 1: %s\" % ceo_ref1[ceo_ref1.plotid.duplicated()].shape[0]\n", + " )\n", + " print(\n", + " \" Number of duplicated in set 2: %s\" % ceo_ref2[ceo_ref2.plotid.duplicated()].shape[0]\n", + " )\n", + " print(\"Removing duplicates and keeping the first...\")\n", + " ceo_ref1 = ceo_ref1.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", + " ceo_ref2 = ceo_ref2.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", + "\n", + " ceo_ref1.set_index(\"plotid\", inplace=True)\n", + " ceo_ref2.set_index(\"plotid\", inplace=True)\n", + " else:\n", + " print(\"The number of rows in the reference sets are equal.\")\n", + "\n", + " ceo_agree = ceo_ref1[ceo_ref1[label_question] == ceo_ref2[label_question]]\n", + "\n", + " print(\n", + " \"Number of samples that are in agreement: %d out of %d (%.2f%%)\"\n", + " % (\n", + " ceo_agree.shape[0],\n", + " ceo_ref1.shape[0],\n", + " ceo_agree.shape[0] / ceo_ref1.shape[0] * 100,\n", + " )\n", + " )\n", + " ceo_agree_geom = gpd.GeoDataFrame(\n", + " ceo_agree,\n", + " geometry=gpd.points_from_xy(ceo_agree.lon, ceo_agree.lat),\n", + " crs=\"EPSG:4326\",\n", + " )\n", + "\n", + " label_responses = ceo_agree_geom[label_question].unique()\n", + " assert len(label_responses) == 2\n", + "\n", + " for r, row in ceo_agree_geom.iterrows():\n", + "\n", + " try:\n", + " if (\n", + " row[label_question].lower() == \"crop\"\n", + " or row[label_question].lower() == \"cropland\"\n", + " or row[label_question].lower() == \"planted\"\n", + " ):\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 1\n", + " elif(\n", + " row[label_question].lower() == \"non-crop\"\n", + " or row[label_question].lower() == \"non-cropland\"\n", + " or row[label_question].lower() == \"not planted\"\n", + " ):\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 0\n", + " except IndexError:\n", + " ceo_agree_geom.loc[r, CLASS_COL] = 255\n", + "\n", + " ceo_agree_geom = ceo_agree_geom[ceo_agree_geom[CLASS_COL] != 255]\n", + "\n", + " ceo_agree_geom[CLASS_COL] = ceo_agree_geom[CLASS_COL].astype(int)\n", + " ceo_agree_geom[COUNTRY_COL] = country\n", + " ceo_agree_geom = ceo_agree_geom[['lat','lon',CLASS_COL, COUNTRY_COL, 'geometry']]\n", + "\n", + " return ceo_agree_geom" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "24QIyHfcZOeG", + "outputId": "17236484-162f-45db-a50f-e205d615f46b" + }, + "id": "24QIyHfcZOeG", + "execution_count": 8, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "gdf = reference_sample_agree(ceo_set1,ceo_set2)\n", + "gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 88 + }, + "id": "QXMdHSHVauqV", + "outputId": "a003c729-6d8f-47d8-827d-ad62206c680b" + }, + "id": "QXMdHSHVauqV", + "execution_count": 9, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Number of NANs/ missing answers in set 1: 2\n", + "Number of NANs/ missing answers in set 2: 0\n", + "The number of rows in the reference sets are equal.\n", + "Number of samples that are in agreement: 487 out of 544 (89.52%)\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "id": "vbVX8gFd_N3J" + }, + "outputs": [], + "source": [ + "!dvc pull data/datasets" + ], + "id": "vbVX8gFd_N3J" + }, + { + "cell_type": "code", + "source": [ + "if not Path(dataset_path).exists():\n", + " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", + "else:\n", + " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", + " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", + " # use only test data because validation points used for harvest-dev map\n", + " df = df[df[\"subset\"] == \"testing\"].copy()\n", + " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", + " df[COUNTRY_COL] = country\n", + "\n", + " gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", + " gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "V8XeT-qci7VG", + "outputId": "042c1313-b2fc-4cc2-9bc1-b26a70b9fe7d" + }, + "id": "V8XeT-qci7VG", + "execution_count": 76, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "markdown", + "metadata": { + "id": "31341d98" + }, + "source": [ + "## 3. Run intercomparison" + ], + "id": "31341d98" + }, + { + "cell_type": "code", + "execution_count": 50, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "ImkKe6cEB4aB", + "outputId": "edea0bbc-2003-429e-cd96-0f6d16acbe40" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " lat lon class_probability subset binary country \\\n", + "0 15.033306 -16.937735 0.000000 testing 0 Senegal \n", + "4 14.799744 -15.329750 0.000000 testing 0 Senegal \n", + "5 14.260755 -14.656014 0.333333 testing 0 Senegal \n", + "6 16.255015 -15.212969 0.000000 testing 0 Senegal \n", + "10 16.201116 -16.003487 0.000000 testing 0 Senegal \n", + "\n", + " geometry \n", + "0 POINT (-16.93773 15.03331) \n", + "4 POINT (-15.32975 14.79974) \n", + "5 POINT (-14.65601 14.26076) \n", + "6 POINT (-15.21297 16.25502) \n", + "10 POINT (-16.00349 16.20112) " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
latlonclass_probabilitysubsetbinarycountrygeometry
015.033306-16.9377350.000000testing0SenegalPOINT (-16.93773 15.03331)
414.799744-15.3297500.000000testing0SenegalPOINT (-15.32975 14.79974)
514.260755-14.6560140.333333testing0SenegalPOINT (-14.65601 14.26076)
616.255015-15.2129690.000000testing0SenegalPOINT (-15.21297 16.25502)
1016.201116-16.0034870.000000testing0SenegalPOINT (-16.00349 16.20112)
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "gdf", + "summary": "{\n \"name\": \"gdf\",\n \"rows\": 602,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1601600883487855,\n \"min\": 12.33836043,\n \"max\": 16.64129064,\n \"num_unique_values\": 322,\n \"samples\": [\n 14.52126657,\n 14.9344916,\n 13.15582734\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.389493345647867,\n \"min\": -17.13536404,\n \"max\": -11.39512938,\n \"num_unique_values\": 371,\n \"samples\": [\n -12.96718113,\n -15.1051715,\n -14.66499701\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.3040067467901026,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 4,\n \"samples\": [\n 0.3333333333333333,\n 0.6666666666666666,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Senegal\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 602,\n \"samples\": [\n \"POINT (-13.58701867 16.09331832)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 50 + } + ], + "source": [ + "gdf.head()" + ], + "id": "ImkKe6cEB4aB" + }, + { + "cell_type": "code", + "source": [ + "gdf2.head()" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 206 + }, + "id": "H1JeTjxdjTNb", + "outputId": "527512fb-7321-4816-ff26-cfc167a1d162" + }, + "id": "H1JeTjxdjTNb", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " lat lon class_probability subset binary country \\\n", + "4 0.725122 34.386282 0.0 testing 0 Kenya \n", + "14 0.621939 34.466496 0.0 testing 0 Kenya \n", + "46 0.226497 34.054859 0.0 testing 0 Kenya \n", + "53 0.096834 33.951959 0.0 testing 0 Kenya \n", + "62 0.032072 33.968469 0.0 testing 0 Kenya \n", + "\n", + " geometry \n", + "4 POINT (34.38628 0.72512) \n", + "14 POINT (34.46650 0.62194) \n", + "46 POINT (34.05486 0.22650) \n", + "53 POINT (33.95196 0.09683) \n", + "62 POINT (33.96847 0.03207) " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
latlonclass_probabilitysubsetbinarycountrygeometry
40.72512234.3862820.0testing0KenyaPOINT (34.38628 0.72512)
140.62193934.4664960.0testing0KenyaPOINT (34.46650 0.62194)
460.22649734.0548590.0testing0KenyaPOINT (34.05486 0.22650)
530.09683433.9519590.0testing0KenyaPOINT (33.95196 0.09683)
620.03207233.9684690.0testing0KenyaPOINT (33.96847 0.03207)
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "variable_name": "gdf2", + "summary": "{\n \"name\": \"gdf2\",\n \"rows\": 764,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.9638166355465471,\n \"min\": -4.10494425,\n \"max\": 3.87641608,\n \"num_unique_values\": 763,\n \"samples\": [\n 0.39123,\n -0.86744,\n 0.476019\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.341905392717519,\n \"min\": 33.95195871,\n \"max\": 41.43769638,\n \"num_unique_values\": 764,\n \"samples\": [\n 34.37571,\n 34.65743,\n 34.23044\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.4347982228847419,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 2,\n \"samples\": [\n 1.0,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 764,\n \"samples\": [\n \"POINT (34.37571 0.39123)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 38 + } + ] + }, + { + "cell_type": "code", + "source": [ + "join_gdf = pd.concat([gdf, gdf2])\n", + "join_gdf.drop(columns=['class_probability', 'subset'])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 424 + }, + "id": "uL0EFpk5jC5M", + "outputId": "3615e7b8-39b6-4635-8104-d829cf2398f0" + }, + "id": "uL0EFpk5jC5M", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " lat lon binary country geometry\n", + "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", + "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", + "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", + "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", + "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)\n", + "... ... ... ... ... ...\n", + "8594 0.633255 34.313450 1 Kenya POINT (34.31345 0.63326)\n", + "8595 0.630631 34.314150 1 Kenya POINT (34.31415 0.63063)\n", + "8603 0.625610 34.307610 1 Kenya POINT (34.30761 0.62561)\n", + "8632 0.694220 34.369890 1 Kenya POINT (34.36989 0.69422)\n", + "8657 0.714165 34.346050 1 Kenya POINT (34.34605 0.71417)\n", + "\n", + "[1251 rows x 5 columns]" + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
latlonbinarycountrygeometry
02.40355334.9886300KenyaPOINT (34.98863 2.40355)
10.23745240.4369130KenyaPOINT (40.43691 0.23745)
24.20326135.5257850KenyaPOINT (35.52579 4.20326)
31.78613335.3450290KenyaPOINT (35.34503 1.78613)
4-2.08870741.0929161KenyaPOINT (41.09292 -2.08871)
..................
85940.63325534.3134501KenyaPOINT (34.31345 0.63326)
85950.63063134.3141501KenyaPOINT (34.31415 0.63063)
86030.62561034.3076101KenyaPOINT (34.30761 0.62561)
86320.69422034.3698901KenyaPOINT (34.36989 0.69422)
86570.71416534.3460501KenyaPOINT (34.34605 0.71417)
\n", + "

1251 rows × 5 columns

\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"join_gdf\",\n \"rows\": 1251,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.411674786387463,\n \"min\": -4.3996825162698165,\n \"max\": 4.509081535107014,\n \"num_unique_values\": 1250,\n \"samples\": [\n 0.05838,\n 0.34771249,\n -3.0539655940358257\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.888586934802481,\n \"min\": 33.95195871,\n \"max\": 41.80068122597383,\n \"num_unique_values\": 1251,\n \"samples\": [\n 35.18629,\n 34.27732,\n 34.16571708\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 1251,\n \"samples\": [\n \"POINT (35.18629 -0.71086)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 40 + } + ] + }, + { + "cell_type": "code", + "execution_count": 72, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "54c4cc0f", + "outputId": "7aea5cb4-d39f-4b27-e725-c2a34e101092" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "TARGETS = {k:v for k,v in TARGETS.items()}\n", + "for k, v in TARGETS.items():\n", + " if country not in v.countries:\n", + " continue\n", + " if v.year is None:\n", + " v.year = v.collection_years[v.countries.index(country)]" + ], + "id": "54c4cc0f" + }, + { + "cell_type": "code", + "execution_count": 73, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "1oQjubrHjkBi", + "outputId": "9c46fc84-6876-4e09-a4a6-26e1ab3cac71" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "reference_year = 2022\n", + "TARGETS = {k: v for k, v in TARGETS.items() if v.year in range(reference_year - 2, reference_year + 1)}\n", + "# TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 2, reference_year, reference_year + 2]}" + ], + "id": "1oQjubrHjkBi" + }, + { + "cell_type": "code", + "source": [ + "for a in range(reference_year - 2, reference_year +1):\n", + " print(a)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 71 + }, + "id": "x6FErLuccwQH", + "outputId": "5c9f08c8-3349-4ded-b491-5e8b75e8fb95" + }, + "id": "x6FErLuccwQH", + "execution_count": 79, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "2020\n", + "2021\n", + "2022\n" + ] + } + ] + }, + { + "cell_type": "code", + "execution_count": 77, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 71 + }, + "id": "98e241d2", + "outputId": "28218d50-cc74-4703-b225-9ccde629d5e9" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[Senegal] sampling worldcover-v100...\n", + "[Senegal] sampling worldcover-v200...\n", + "[Senegal] sampling worldcereal-v100...\n" + ] + } + ], + "source": [ + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", + " map_sampled = cropmap.extract_test(gdf).copy()\n", + " gdf = pd.merge(gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", + " gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" + ], + "id": "98e241d2" + }, + { + "cell_type": "code", + "source": [ + "# for cropmap in TARGETS.values():\n", + "# if country not in cropmap.countries:\n", + "# continue\n", + "# print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", + "# map_sampled = cropmap.extract_test(join_gdf).copy()\n", + "# join_gdf = pd.merge(join_gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", + "# join_gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "l9WMBBIOjsRS", + "outputId": "15d68bc1-33da-4cab-91c8-ecebfcd5dbe6" + }, + "id": "l9WMBBIOjsRS", + "execution_count": 13, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 81, + "metadata": { + "id": "95a0f536", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 124 + }, + "outputId": "838855c1-13ed-4940-b0ce-4c106a80f7c1" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[Senegal] calculating pixel area for worldcover-v100...\n", + "Export task started for worldcover-v100, Senegal. Returning null for now.\n", + "[Senegal] calculating pixel area for worldcover-v200...\n", + "Export task started for worldcover-v200, Senegal. Returning null for now.\n", + "[Senegal] calculating pixel area for worldcereal-v100...\n", + "Export task started for worldcereal-v100, Senegal. Returning null for now.\n" + ] + } + ], + "source": [ + "a_j = {}\n", + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", + " a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", + " # a_j[cropmap.title] = cropmap.compute_map_area(country, dataset_name=cropmap.title).copy()\n", + " # a_j[cropmap.title] = np.array([None,None])\n" + ], + "id": "95a0f536" + }, + { + "cell_type": "code", + "source": [ + "# update a_j values with exported values\n", + "for cropmap in a_j.keys():\n", + " try:\n", + " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-{country}-{cropmap}.csv')\n", + " except:\n", + " continue\n", + " crop_area = int(area_df['crop_sum'][0])\n", + " noncrop_area = int(area_df['noncrop_sum'][0])\n", + " a_j[cropmap] = np.array([noncrop_area, crop_area])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "5fJPzvOeUo9G", + "outputId": "3377fa7c-f455-46ab-9b36-5e9b2c2101d4" + }, + "id": "5fJPzvOeUo9G", + "execution_count": 83, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "# Change None to nan\n", + "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "zyR4qCJ49Rh5", + "outputId": "426a7d13-545a-4261-f84a-42e68f6b9409" + }, + "id": "zyR4qCJ49Rh5", + "execution_count": 85, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 86, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "LY6Q_QtUgME_", + "outputId": "5839d191-f940-48cf-96f5-2f95900ce4f2" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "from src.area_utils import compute_area_estimate, compute_area_error_matrix, compute_std_p_i\n", + "from sklearn.metrics import confusion_matrix" + ], + "id": "LY6Q_QtUgME_" + }, + { + "cell_type": "code", + "execution_count": 87, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "oojPqwSboiWU", + "outputId": "091f0d2b-2c73-42af-8055-e28ca741ef2c" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "def compute_area_estimate(dataset, true, pred, a_j, resolution):\n", + " cm = confusion_matrix(true, pred)\n", + " total_px = a_j.sum()\n", + " w_j = a_j / total_px\n", + "\n", + " am = compute_area_error_matrix(cm, w_j)\n", + " a_i = am.sum(axis=1)\n", + " std_a_i = compute_std_p_i(w_j, am, cm)\n", + " err_a_i = 1.96 * std_a_i\n", + "\n", + " a_px = total_px * a_i\n", + " err_px = err_a_i * total_px\n", + " return pd.DataFrame(\n", + " data={\n", + " \"dataset\": dataset,\n", + " \"area_ha\": a_px[1] * (resolution**2) / (100**2),\n", + " \"err_ha\": err_px[1] * (resolution**2) / (100**2),\n", + " },\n", + " index=[0],\n", + " ).round(2)" + ], + "id": "oojPqwSboiWU" + }, + { + "cell_type": "code", + "execution_count": 111, + "metadata": { + "id": "ti5ZXmbyn6Mm", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "outputId": "10b229e0-28ff-4690-d571-ce286ac789df" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "comparisons = []\n", + "area_est = []\n", + "for cropmap in TARGETS.values():\n", + " cropmap, resolution = cropmap.title, cropmap.resolution\n", + " if cropmap not in gdf.columns:\n", + " continue\n", + " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", + " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", + " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", + " comparisons.append(comparison)\n", + " area_est.append(area)\n", + "\n", + "# # Add ensemble\n", + "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", + "\n", + "# print(f\"Ensemble maps: {ensemble_maps}\")\n", + "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", + "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", + "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", + "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", + "# comparisons.append(comparison)\n", + "# area_est.append(area)\n", + "\n", + "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", + "area_est = pd.concat(area_est).set_index(['dataset'])\n", + "\n", + "results = comparisons.merge(area_est, on='dataset')" + ], + "id": "ti5ZXmbyn6Mm" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "QrAgv7pP1lcz", + "outputId": "6f33c955-6ceb-4295-84ed-4aaf65c1512f" + }, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + " \n", + " " + ], + "text/plain": [ + "" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "results.to_csv('results.csv')" + ], + "id": "QrAgv7pP1lcz" + }, + { + "cell_type": "code", + "source": [ + "results.columns" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 124 + }, + "id": "xOO6fdt0CiG6", + "outputId": "5ba33666-7b75-4785-915c-5a2cb3a7d12f" + }, + "id": "xOO6fdt0CiG6", + "execution_count": null, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + "Index(['country', 'crop_f1', 'std_crop_f1', 'accuracy', 'std_acc',\n", + " 'crop_recall_pa', 'std_crop_pa', 'noncrop_recall_pa', 'std_noncrop_pa',\n", + " 'crop_precision_ua', 'std_crop_ua', 'noncrop_precision_ua',\n", + " 'std_noncrop_ua', 'crop_support', 'noncrop_support', 'tn', 'fp', 'fn',\n", + " 'tp', 'tn_area', 'fp_area', 'fn_area', 'tp_area', 'area_ha', 'err_ha'],\n", + " dtype='object')" + ] + }, + "metadata": {}, + "execution_count": 21 + } + ] + }, + { + "cell_type": "code", + "execution_count": 89, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 175 + }, + "id": "nAj0p7VS1_2K", + "outputId": "e2f2e6fd-542f-4a26-89e7-0cc017cd4753" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", + "dataset \n", + "worldcover-v100 0.64 0.88 0.01 0.66 0.04 \n", + "worldcover-v200 0.65 0.89 0.01 0.69 0.04 \n", + "worldcereal-v100 0.64 0.88 0.01 0.66 0.04 \n", + "\n", + " crop_precision_ua std_crop_ua area_ha err_ha \n", + "dataset \n", + "worldcover-v100 0.61 0.05 3097982.37 486283.35 \n", + "worldcover-v200 0.62 0.05 3068403.67 475012.65 \n", + "worldcereal-v100 0.62 0.05 3184080.79 495468.16 " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
crop_f1accuracystd_acccrop_recall_pastd_crop_pacrop_precision_uastd_crop_uaarea_haerr_ha
dataset
worldcover-v1000.640.880.010.660.040.610.053097982.37486283.35
worldcover-v2000.650.890.010.690.040.620.053068403.67475012.65
worldcereal-v1000.640.880.010.660.040.620.053184080.79495468.16
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"dataset\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"worldcover-v100\",\n \"worldcover-v200\",\n \"worldcereal-v100\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_f1\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896263,\n \"min\": 0.64,\n \"max\": 0.65,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.65,\n 0.64\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"accuracy\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896262,\n \"min\": 0.88,\n \"max\": 0.89,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.89,\n 0.88\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_acc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.01,\n \"max\": 0.01,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_recall_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.017320508075688724,\n \"min\": 0.66,\n \"max\": 0.69,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.69\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.04,\n \"max\": 0.04,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.04\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_precision_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896262,\n \"min\": 0.61,\n \"max\": 0.62,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.62\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8.498374721940739e-18,\n \"min\": 0.05,\n \"max\": 0.05,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.05\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"area_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 60095.80339158582,\n \"min\": 3068403.67,\n \"max\": 3184080.79,\n \"num_unique_values\": 3,\n \"samples\": [\n 3097982.37\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"err_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 10245.46487139714,\n \"min\": 475012.65,\n \"max\": 495468.16,\n \"num_unique_values\": 3,\n \"samples\": [\n 486283.35\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 89 + } + ], + "source": [ + "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI" + ], + "id": "nAj0p7VS1_2K" + }, + { + "cell_type": "markdown", + "metadata": { + "id": "fa969373" + }, + "source": [ + "## 4. Visualize best available map" + ], + "id": "fa969373" + }, + { + "cell_type": "code", + "source": [ + "results.dropna(inplace=True)" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "qenOtnORfGTR", + "outputId": "e4430a49-6836-44be-b753-86f58ac8e387" + }, + "id": "qenOtnORfGTR", + "execution_count": 29, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 90, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "fraQjcTMpTwp", + "outputId": "4b4b7209-c8c7-4edc-9c67-3c8145bf61cd" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "crop_proportion = round(gdf[CLASS_COL].value_counts(normalize=True)[1], 4) * 100\n", + "ax = results.sort_values(\"crop_f1\").plot(\n", + " y=[\"accuracy\", \"crop_recall_pa\", \"crop_precision_ua\", \"crop_f1\"],\n", + " xerr=\"std_crop_f1\",\n", + " kind=\"barh\",\n", + " figsize=(6, 14),\n", + " width=0.8,\n", + " title=f\"{country}: {len(gdf)} points (crop proportion: {crop_proportion}%)\",\n", + ");\n", + "\n", + "for c in ax.containers[1::2]:\n", + " ax.bar_label(c)\n", + "\n", + "for border in [\"top\", \"right\", \"bottom\", \"left\"]:\n", + " ax.spines[border].set_visible(False)\n", + "\n", + "ax.legend(bbox_to_anchor=(1, 1), reverse=True);" + ], + "id": "fraQjcTMpTwp" + }, + { + "cell_type": "code", + "source": [ + "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", + "# fao_stat = fao_stat[fao_stat['Area'] == country]\n", + "fao_stat = fao_stat[fao_stat['Area'] == country]['Value'].mean() * 1000\n", + "# fao_stat = fao_stat[fao_stat['Year Code'] == reference_year]['Value'] * 1000\n", + "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "L-nrhBekPfcp", + "outputId": "bf29eab8-ff43-4d88-983a-6902728c5eec" + }, + "id": "L-nrhBekPfcp", + "execution_count": 116, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [ + "fao_stat[fao_stat['Area'] == country]" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 175 + }, + "id": "rF9DADLtnX4B", + "outputId": "a5191569-eb17-438c-ac3a-d8ae63c7c449" + }, + "id": "rF9DADLtnX4B", + "execution_count": 115, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "execute_result", + "data": { + "text/plain": [ + " Domain Code Domain Area Code (M49) Area Element Code Element \\\n", + "56 RL Land Use 686 Senegal 5110 Area \n", + "57 RL Land Use 686 Senegal 5110 Area \n", + "58 RL Land Use 686 Senegal 5110 Area \n", + "59 RL Land Use 686 Senegal 5110 Area \n", + "\n", + " Item Code Item Year Code Year Unit Value Flag \\\n", + "56 6620 Cropland 2018 2018 1000 ha 3758.0 I \n", + "57 6620 Cropland 2019 2019 1000 ha 3795.0 I \n", + "58 6620 Cropland 2020 2020 1000 ha 3830.0 I \n", + "59 6620 Cropland 2021 2021 1000 ha 3911.0 I \n", + "\n", + " Flag Description Note \n", + "56 Imputed value NaN \n", + "57 Imputed value NaN \n", + "58 Imputed value NaN \n", + "59 Imputed value NaN " + ], + "text/html": [ + "\n", + "
\n", + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
Domain CodeDomainArea Code (M49)AreaElement CodeElementItem CodeItemYear CodeYearUnitValueFlagFlag DescriptionNote
56RLLand Use686Senegal5110Area6620Cropland201820181000 ha3758.0IImputed valueNaN
57RLLand Use686Senegal5110Area6620Cropland201920191000 ha3795.0IImputed valueNaN
58RLLand Use686Senegal5110Area6620Cropland202020201000 ha3830.0IImputed valueNaN
59RLLand Use686Senegal5110Area6620Cropland202120211000 ha3911.0IImputed valueNaN
\n", + "
\n", + "
\n", + "\n", + "
\n", + " \n", + "\n", + " \n", + "\n", + " \n", + "
\n", + "\n", + "\n", + "
\n", + " \n", + "\n", + "\n", + "\n", + " \n", + "
\n", + "
\n", + "
\n" + ], + "application/vnd.google.colaboratory.intrinsic+json": { + "type": "dataframe", + "summary": "{\n \"name\": \"fao_stat[fao_stat['Area'] == country]\",\n \"rows\": 4,\n \"fields\": [\n {\n \"column\": \"Domain Code\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"RL\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Domain\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Land Use\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Area Code (M49)\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 686,\n \"max\": 686,\n \"num_unique_values\": 1,\n \"samples\": [\n 686\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Area\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Senegal\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Element Code\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 5110,\n \"max\": 5110,\n \"num_unique_values\": 1,\n \"samples\": [\n 5110\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Element\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Area\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Item Code\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 6620,\n \"max\": 6620,\n \"num_unique_values\": 1,\n \"samples\": [\n 6620\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Item\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Cropland\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Year Code\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1,\n \"min\": 2018,\n \"max\": 2021,\n \"num_unique_values\": 4,\n \"samples\": [\n 2019\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Year\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1,\n \"min\": 2018,\n \"max\": 2021,\n \"num_unique_values\": 4,\n \"samples\": [\n 2019\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Unit\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"1000 ha\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Value\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 65.32227797620043,\n \"min\": 3758.0,\n \"max\": 3911.0,\n \"num_unique_values\": 4,\n \"samples\": [\n 3795.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Flag\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"I\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Flag Description\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Imputed value\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"Note\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": null,\n \"min\": null,\n \"max\": null,\n \"num_unique_values\": 0,\n \"samples\": [],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" + } + }, + "metadata": {}, + "execution_count": 115 + } + ] + }, + { + "cell_type": "code", + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import matplotlib.pyplot as plt\n", + "fig, ax = plt.subplots()\n", + "\n", + "n = len(results)\n", + "colors = plt.cm.viridis(np.linspace(0, 1, n))\n", + "\n", + "ax.barh(\n", + " results.index,\n", + " results[\"area_ha\"],\n", + " xerr=results[\"err_ha\"],\n", + " align=\"center\",\n", + " alpha=0.5,\n", + " ecolor=\"black\",\n", + " color= colors\n", + ")\n", + "\n", + "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", + " ax.text(value, i, f\"{value:,} ± {err:,}\", ha=\"center\", va=\"bottom\")\n", + "ax.set_ylabel(\"Area (ha)\")\n", + "ax.set_title(\"Area of cropland\")\n", + "ax.spines[\"right\"].set_visible(False)\n", + "plt.show()" + ], + "metadata": { + "id": "a0XEODxnBXW3", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 470 + }, + "outputId": "0be8683a-47d8-4137-9530-447be6bbd9de" + }, + "id": "a0XEODxnBXW3", + "execution_count": 110, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "source": [], + "metadata": { + "id": "gfTIBg6cwwAZ" + }, + "id": "gfTIBg6cwwAZ", + "execution_count": null, + "outputs": [] + } + ], + "metadata": { + "colab": { + "provenance": [] + }, + "kernelspec": { + "display_name": "Python 3 (ipykernel)", + "language": "python", + "name": "python3" + }, + "language_info": { + "codemirror_mode": { + "name": "ipython", + "version": 3 + }, + "file_extension": ".py", + "mimetype": "text/x-python", + "name": "python", + "nbconvert_exporter": "python", + "pygments_lexer": "ipython3", + "version": "3.8.15" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +} \ No newline at end of file From 6590ed312936c24ee87670f9d5caf4964c78933c Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Sat, 23 Mar 2024 23:33:20 +0000 Subject: [PATCH 12/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index 2cbf7b85..51887fd5 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -3689,4 +3689,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From fc7691ce28ba984ebb346891f55f3a14b56b7350 Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Sat, 23 Mar 2024 19:48:31 -0400 Subject: [PATCH 13/21] Update Senegal area estimates notebook --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 3136 ++++------------- 1 file changed, 753 insertions(+), 2383 deletions(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index 51887fd5..3aa84cb3 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -38,7 +38,7 @@ "colab": { "base_uri": "https://localhost:8080/" }, - "outputId": "80951fe8-fd11-4fd3-ce7a-f92012496b2d" + "outputId": "7ae24c2f-c6e5-4832-8b23-a00a30836ce2" }, "outputs": [ { @@ -46,12 +46,12 @@ "name": "stdout", "text": [ "Cloning into 'crop-mask'...\n", - "remote: Enumerating objects: 12074, done.\u001b[K\n", + "remote: Enumerating objects: 12121, done.\u001b[K\n", "remote: Counting objects: 100% (1485/1485), done.\u001b[K\n", - "remote: Compressing objects: 100% (449/449), done.\u001b[K\n", - "remote: Total 12074 (delta 1102), reused 1232 (delta 1009), pack-reused 10589\u001b[K\n", - "Receiving objects: 100% (12074/12074), 125.43 MiB | 11.56 MiB/s, done.\n", - "Resolving deltas: 100% (7824/7824), done.\n", + "remote: Compressing objects: 100% (472/472), done.\u001b[K\n", + "remote: Total 12121 (delta 1100), reused 1215 (delta 986), pack-reused 10636\u001b[K\n", + "Receiving objects: 100% (12121/12121), 125.63 MiB | 8.99 MiB/s, done.\n", + "Resolving deltas: 100% (7861/7861), done.\n", "Updating files: 100% (208/208), done.\n" ] } @@ -69,7 +69,7 @@ "base_uri": "https://localhost:8080/" }, "id": "1fe-6D3f8LTb", - "outputId": "6c6848be-2e5f-4c10-ce9c-b4dc071a2795" + "outputId": "95bcc387-eb42-4921-afb5-193377e147af" }, "outputs": [ { @@ -95,7 +95,7 @@ "base_uri": "https://localhost:8080/" }, "id": "V6lTs8Z9Pt-T", - "outputId": "a9ed0471-9de0-4299-b537-069aa07a453c" + "outputId": "9157afbd-b0be-4978-bd00-28bd74e9d17a" }, "id": "V6lTs8Z9Pt-T", "execution_count": 3, @@ -126,49 +126,23 @@ }, { "cell_type": "code", - "execution_count": 71, + "execution_count": 5, "metadata": { "id": "9907f9a5", "colab": { "base_uri": "https://localhost:8080/", - "height": 17 + "height": 73 }, - "outputId": "4e49e10a-1dc2-44f2-f39f-0acb74d3845b" + "outputId": "762ee4ed-a169-43d5-968e-01e71f287bf3" }, "outputs": [ { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} + "output_type": "stream", + "name": "stderr", + "text": [ + "/usr/local/lib/python3.10/dist-packages/cartopy/io/__init__.py:241: DownloadWarning: Downloading: https://naturalearth.s3.amazonaws.com/10m_cultural/ne_10m_admin_1_states_provinces.zip\n", + " warnings.warn(f'Downloading: {url}', DownloadWarning)\n" + ] } ], "source": [ @@ -203,14 +177,14 @@ }, { "cell_type": "code", - "execution_count": 48, + "execution_count": 6, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 17 }, "id": "7f75e567", - "outputId": "ccea9d96-7484-4157-9cee-2058c97bbc02" + "outputId": "09fdc14d-3af8-4002-f7fb-daf01e212d62" }, "outputs": [ { @@ -265,14 +239,14 @@ }, { "cell_type": "code", - "execution_count": 35, + "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 17 }, "id": "prvHkUXTOe7P", - "outputId": "fe22a240-697a-4986-bfa7-f049935bd1cc" + "outputId": "3fc18281-33d5-4d2e-b371-5b90d054f990" }, "outputs": [ { @@ -318,251 +292,26 @@ }, { "cell_type": "code", - "source": [ - "# ceo_set1 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv'\n", - "# ceo_set2 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv'" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "66-YJBNxYAdF", - "outputId": "221a20ee-9808-4181-8354-d1877c544aca" - }, - "id": "66-YJBNxYAdF", - "execution_count": 7, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "source": [ - "def reference_sample_agree(ceo_ref1, ceo_ref2):\n", - " ceo_ref1 = pd.read_csv(ceo_ref1)\n", - " ceo_ref2 = pd.read_csv(ceo_ref2)\n", - "\n", - " assert ceo_ref1.columns[-1] == ceo_ref2.columns[-1]\n", - "\n", - " label_question = ceo_ref1.columns[-1]\n", - "\n", - " print(f\"Number of NANs/ missing answers in set 1: {ceo_ref1[label_question].isna().sum()}\")\n", - " print(f\"Number of NANs/ missing answers in set 2: {ceo_ref2[label_question].isna().sum()}\")\n", - "\n", - " if ceo_ref1.shape[0] != ceo_ref2.shape[0]:\n", - " print(\"The number of rows in the reference sets are not equal.\")\n", - " print(\"Checking for duplictes on 'plotid'..\")\n", - " print(\n", - " \" Number of duplicated in set 1: %s\" % ceo_ref1[ceo_ref1.plotid.duplicated()].shape[0]\n", - " )\n", - " print(\n", - " \" Number of duplicated in set 2: %s\" % ceo_ref2[ceo_ref2.plotid.duplicated()].shape[0]\n", - " )\n", - " print(\"Removing duplicates and keeping the first...\")\n", - " ceo_ref1 = ceo_ref1.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", - " ceo_ref2 = ceo_ref2.drop_duplicates(subset=\"plotid\", keep=\"first\")\n", - "\n", - " ceo_ref1.set_index(\"plotid\", inplace=True)\n", - " ceo_ref2.set_index(\"plotid\", inplace=True)\n", - " else:\n", - " print(\"The number of rows in the reference sets are equal.\")\n", - "\n", - " ceo_agree = ceo_ref1[ceo_ref1[label_question] == ceo_ref2[label_question]]\n", - "\n", - " print(\n", - " \"Number of samples that are in agreement: %d out of %d (%.2f%%)\"\n", - " % (\n", - " ceo_agree.shape[0],\n", - " ceo_ref1.shape[0],\n", - " ceo_agree.shape[0] / ceo_ref1.shape[0] * 100,\n", - " )\n", - " )\n", - " ceo_agree_geom = gpd.GeoDataFrame(\n", - " ceo_agree,\n", - " geometry=gpd.points_from_xy(ceo_agree.lon, ceo_agree.lat),\n", - " crs=\"EPSG:4326\",\n", - " )\n", - "\n", - " label_responses = ceo_agree_geom[label_question].unique()\n", - " assert len(label_responses) == 2\n", - "\n", - " for r, row in ceo_agree_geom.iterrows():\n", - "\n", - " try:\n", - " if (\n", - " row[label_question].lower() == \"crop\"\n", - " or row[label_question].lower() == \"cropland\"\n", - " or row[label_question].lower() == \"planted\"\n", - " ):\n", - " ceo_agree_geom.loc[r, CLASS_COL] = 1\n", - " elif(\n", - " row[label_question].lower() == \"non-crop\"\n", - " or row[label_question].lower() == \"non-cropland\"\n", - " or row[label_question].lower() == \"not planted\"\n", - " ):\n", - " ceo_agree_geom.loc[r, CLASS_COL] = 0\n", - " except IndexError:\n", - " ceo_agree_geom.loc[r, CLASS_COL] = 255\n", - "\n", - " ceo_agree_geom = ceo_agree_geom[ceo_agree_geom[CLASS_COL] != 255]\n", - "\n", - " ceo_agree_geom[CLASS_COL] = ceo_agree_geom[CLASS_COL].astype(int)\n", - " ceo_agree_geom[COUNTRY_COL] = country\n", - " ceo_agree_geom = ceo_agree_geom[['lat','lon',CLASS_COL, COUNTRY_COL, 'geometry']]\n", - "\n", - " return ceo_agree_geom" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "24QIyHfcZOeG", - "outputId": "17236484-162f-45db-a50f-e205d615f46b" - }, - "id": "24QIyHfcZOeG", "execution_count": 8, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "source": [ - "gdf = reference_sample_agree(ceo_set1,ceo_set2)\n", - "gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" - ], "metadata": { + "id": "vbVX8gFd_N3J", + "outputId": "f943aa0f-f6b4-437c-a70f-6eed2c07e76e", "colab": { - "base_uri": "https://localhost:8080/", - "height": 88 - }, - "id": "QXMdHSHVauqV", - "outputId": "a003c729-6d8f-47d8-827d-ad62206c680b" + "base_uri": "https://localhost:8080/" + } }, - "id": "QXMdHSHVauqV", - "execution_count": 9, "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, { "output_type": "stream", "name": "stdout", "text": [ - "Number of NANs/ missing answers in set 1: 2\n", - "Number of NANs/ missing answers in set 2: 0\n", - "The number of rows in the reference sets are equal.\n", - "Number of samples that are in agreement: 487 out of 544 (89.52%)\n" + "Applying changes |52.0 [00:06, 8.27file/s]\n", + "\u001b[32mA\u001b[0m data/datasets/\n", + "1 file added and 53 files fetched\n", + "\u001b[0m" ] } - ] - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "id": "vbVX8gFd_N3J" - }, - "outputs": [], + ], "source": [ "!dvc pull data/datasets" ], @@ -577,7 +326,7 @@ " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", " # use only test data because validation points used for harvest-dev map\n", - " df = df[df[\"subset\"] == \"testing\"].copy()\n", + " df = df[(df[\"subset\"] == \"validation\") | (df[\"subset\"] == \"testing\")].copy()\n", " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", " df[COUNTRY_COL] = country\n", "\n", @@ -590,10 +339,10 @@ "height": 17 }, "id": "V8XeT-qci7VG", - "outputId": "042c1313-b2fc-4cc2-9bc1-b26a70b9fe7d" + "outputId": "68dfc81c-1b81-4dfc-f590-f07e32ba9542" }, "id": "V8XeT-qci7VG", - "execution_count": 76, + "execution_count": 10, "outputs": [ { "output_type": "display_data", @@ -643,14 +392,14 @@ }, { "cell_type": "code", - "execution_count": 50, + "execution_count": 11, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 206 }, "id": "ImkKe6cEB4aB", - "outputId": "edea0bbc-2003-429e-cd96-0f6d16acbe40" + "outputId": "f0df68d8-dec2-485b-80a7-c1d033d66fee" }, "outputs": [ { @@ -691,23 +440,23 @@ "output_type": "execute_result", "data": { "text/plain": [ - " lat lon class_probability subset binary country \\\n", - "0 15.033306 -16.937735 0.000000 testing 0 Senegal \n", - "4 14.799744 -15.329750 0.000000 testing 0 Senegal \n", - "5 14.260755 -14.656014 0.333333 testing 0 Senegal \n", - "6 16.255015 -15.212969 0.000000 testing 0 Senegal \n", - "10 16.201116 -16.003487 0.000000 testing 0 Senegal \n", - "\n", - " geometry \n", - "0 POINT (-16.93773 15.03331) \n", - "4 POINT (-15.32975 14.79974) \n", - "5 POINT (-14.65601 14.26076) \n", - "6 POINT (-15.21297 16.25502) \n", - "10 POINT (-16.00349 16.20112) " + " lat lon class_probability subset binary country \\\n", + "0 15.033306 -16.937735 0.000000 testing 0 Senegal \n", + "2 16.192133 -14.772795 0.000000 validation 0 Senegal \n", + "3 15.015340 -13.173794 0.000000 validation 0 Senegal \n", + "4 14.799744 -15.329750 0.000000 testing 0 Senegal \n", + "5 14.260755 -14.656014 0.333333 testing 0 Senegal \n", + "\n", + " geometry \n", + "0 POINT (-16.93773 15.03331) \n", + "2 POINT (-14.77279 16.19213) \n", + "3 POINT (-13.17379 15.01534) \n", + "4 POINT (-15.32975 14.79974) \n", + "5 POINT (-14.65601 14.26076) " ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
latlonclass_probabilitysubsetbinarycountrygeometry
40.72512234.3862820.0testing0KenyaPOINT (34.38628 0.72512)
140.62193934.4664960.0testing0KenyaPOINT (34.46650 0.62194)
460.22649734.0548590.0testing0KenyaPOINT (34.05486 0.22650)
530.09683433.9519590.0testing0KenyaPOINT (33.95196 0.09683)
620.03207233.9684690.0testing0KenyaPOINT (33.96847 0.03207)
\n", - "
\n", - "
\n", - "\n", - "
\n", - " \n", - "\n", - " \n", - "\n", - " \n", - "
\n", - "\n", - "\n", - "
\n", - " \n", - "\n", - "\n", - "\n", - " \n", - "
\n", - "
\n", - "
\n" - ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "variable_name": "gdf2", - "summary": "{\n \"name\": \"gdf2\",\n \"rows\": 764,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.9638166355465471,\n \"min\": -4.10494425,\n \"max\": 3.87641608,\n \"num_unique_values\": 763,\n \"samples\": [\n 0.39123,\n -0.86744,\n 0.476019\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.341905392717519,\n \"min\": 33.95195871,\n \"max\": 41.43769638,\n \"num_unique_values\": 764,\n \"samples\": [\n 34.37571,\n 34.65743,\n 34.23044\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.4347982228847419,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 2,\n \"samples\": [\n 1.0,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 764,\n \"samples\": [\n \"POINT (34.37571 0.39123)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } - }, - "metadata": {}, - "execution_count": 38 - } - ] - }, - { - "cell_type": "code", - "source": [ - "join_gdf = pd.concat([gdf, gdf2])\n", - "join_gdf.drop(columns=['class_probability', 'subset'])" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 424 - }, - "id": "uL0EFpk5jC5M", - "outputId": "3615e7b8-39b6-4635-8104-d829cf2398f0" - }, - "id": "uL0EFpk5jC5M", - "execution_count": null, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "execute_result", - "data": { - "text/plain": [ - " lat lon binary country geometry\n", - "0 2.403553 34.988630 0 Kenya POINT (34.98863 2.40355)\n", - "1 0.237452 40.436913 0 Kenya POINT (40.43691 0.23745)\n", - "2 4.203261 35.525785 0 Kenya POINT (35.52579 4.20326)\n", - "3 1.786133 35.345029 0 Kenya POINT (35.34503 1.78613)\n", - "4 -2.088707 41.092916 1 Kenya POINT (41.09292 -2.08871)\n", - "... ... ... ... ... ...\n", - "8594 0.633255 34.313450 1 Kenya POINT (34.31345 0.63326)\n", - "8595 0.630631 34.314150 1 Kenya POINT (34.31415 0.63063)\n", - "8603 0.625610 34.307610 1 Kenya POINT (34.30761 0.62561)\n", - "8632 0.694220 34.369890 1 Kenya POINT (34.36989 0.69422)\n", - "8657 0.714165 34.346050 1 Kenya POINT (34.34605 0.71417)\n", - "\n", - "[1251 rows x 5 columns]" - ], - "text/html": [ - "\n", - "
\n", - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
latlonbinarycountrygeometry
02.40355334.9886300KenyaPOINT (34.98863 2.40355)
10.23745240.4369130KenyaPOINT (40.43691 0.23745)
24.20326135.5257850KenyaPOINT (35.52579 4.20326)
31.78613335.3450290KenyaPOINT (35.34503 1.78613)
4-2.08870741.0929161KenyaPOINT (41.09292 -2.08871)
..................
85940.63325534.3134501KenyaPOINT (34.31345 0.63326)
85950.63063134.3141501KenyaPOINT (34.31415 0.63063)
86030.62561034.3076101KenyaPOINT (34.30761 0.62561)
86320.69422034.3698901KenyaPOINT (34.36989 0.69422)
86570.71416534.3460501KenyaPOINT (34.34605 0.71417)
\n", - "

1251 rows × 5 columns

\n", - "
\n", - "
\n", - "\n", - "
\n", - " \n", - "\n", - " \n", - "\n", - " \n", - "
\n", - "\n", - "\n", - "
\n", - " \n", - "\n", - "\n", - "\n", - " \n", - "
\n", - "
\n", - "
\n" - ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "summary": "{\n \"name\": \"join_gdf\",\n \"rows\": 1251,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.411674786387463,\n \"min\": -4.3996825162698165,\n \"max\": 4.509081535107014,\n \"num_unique_values\": 1250,\n \"samples\": [\n 0.05838,\n 0.34771249,\n -3.0539655940358257\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.888586934802481,\n \"min\": 33.95195871,\n \"max\": 41.80068122597383,\n \"num_unique_values\": 1251,\n \"samples\": [\n 35.18629,\n 34.27732,\n 34.16571708\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Kenya\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 1251,\n \"samples\": [\n \"POINT (35.18629 -0.71086)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } - }, - "metadata": {}, - "execution_count": 40 - } - ] - }, - { - "cell_type": "code", - "execution_count": 72, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "54c4cc0f", - "outputId": "7aea5cb4-d39f-4b27-e725-c2a34e101092" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ], - "source": [ - "TARGETS = {k:v for k,v in TARGETS.items()}\n", - "for k, v in TARGETS.items():\n", - " if country not in v.countries:\n", - " continue\n", - " if v.year is None:\n", - " v.year = v.collection_years[v.countries.index(country)]" - ], - "id": "54c4cc0f" - }, - { - "cell_type": "code", - "execution_count": 73, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "1oQjubrHjkBi", - "outputId": "9c46fc84-6876-4e09-a4a6-26e1ab3cac71" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ], - "source": [ - "reference_year = 2022\n", - "TARGETS = {k: v for k, v in TARGETS.items() if v.year in range(reference_year - 2, reference_year + 1)}\n", - "# TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 2, reference_year, reference_year + 2]}" - ], - "id": "1oQjubrHjkBi" - }, - { - "cell_type": "code", - "source": [ - "for a in range(reference_year - 2, reference_year +1):\n", - " print(a)" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 71 - }, - "id": "x6FErLuccwQH", - "outputId": "5c9f08c8-3349-4ded-b491-5e8b75e8fb95" - }, - "id": "x6FErLuccwQH", - "execution_count": 79, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "2020\n", - "2021\n", - "2022\n" - ] - } - ] - }, - { - "cell_type": "code", - "execution_count": 77, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 71 - }, - "id": "98e241d2", - "outputId": "28218d50-cc74-4703-b225-9ccde629d5e9" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "[Senegal] sampling worldcover-v100...\n", - "[Senegal] sampling worldcover-v200...\n", - "[Senegal] sampling worldcereal-v100...\n" - ] - } - ], - "source": [ - "for cropmap in TARGETS.values():\n", - " if country not in cropmap.countries:\n", - " continue\n", - " print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", - " map_sampled = cropmap.extract_test(gdf).copy()\n", - " gdf = pd.merge(gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", - " gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" - ], - "id": "98e241d2" - }, - { - "cell_type": "code", - "source": [ - "# for cropmap in TARGETS.values():\n", - "# if country not in cropmap.countries:\n", - "# continue\n", - "# print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", - "# map_sampled = cropmap.extract_test(join_gdf).copy()\n", - "# join_gdf = pd.merge(join_gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", - "# join_gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "l9WMBBIOjsRS", - "outputId": "15d68bc1-33da-4cab-91c8-ecebfcd5dbe6" - }, - "id": "l9WMBBIOjsRS", - "execution_count": 13, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "execution_count": 81, - "metadata": { - "id": "95a0f536", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 124 - }, - "outputId": "838855c1-13ed-4940-b0ce-4c106a80f7c1" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "stream", - "name": "stdout", - "text": [ - "[Senegal] calculating pixel area for worldcover-v100...\n", - "Export task started for worldcover-v100, Senegal. Returning null for now.\n", - "[Senegal] calculating pixel area for worldcover-v200...\n", - "Export task started for worldcover-v200, Senegal. Returning null for now.\n", - "[Senegal] calculating pixel area for worldcereal-v100...\n", - "Export task started for worldcereal-v100, Senegal. Returning null for now.\n" - ] - } - ], - "source": [ - "a_j = {}\n", - "for cropmap in TARGETS.values():\n", - " if country not in cropmap.countries:\n", - " continue\n", - " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", - " a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", - " # a_j[cropmap.title] = cropmap.compute_map_area(country, dataset_name=cropmap.title).copy()\n", - " # a_j[cropmap.title] = np.array([None,None])\n" - ], - "id": "95a0f536" - }, - { - "cell_type": "code", - "source": [ - "# update a_j values with exported values\n", - "for cropmap in a_j.keys():\n", - " try:\n", - " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-{country}-{cropmap}.csv')\n", - " except:\n", - " continue\n", - " crop_area = int(area_df['crop_sum'][0])\n", - " noncrop_area = int(area_df['noncrop_sum'][0])\n", - " a_j[cropmap] = np.array([noncrop_area, crop_area])" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "5fJPzvOeUo9G", - "outputId": "3377fa7c-f455-46ab-9b36-5e9b2c2101d4" - }, - "id": "5fJPzvOeUo9G", - "execution_count": 83, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "source": [ - "# Change None to nan\n", - "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "zyR4qCJ49Rh5", - "outputId": "426a7d13-545a-4261-f84a-42e68f6b9409" - }, - "id": "zyR4qCJ49Rh5", - "execution_count": 85, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ] - }, - { - "cell_type": "code", - "execution_count": 86, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "LY6Q_QtUgME_", - "outputId": "5839d191-f940-48cf-96f5-2f95900ce4f2" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ], - "source": [ - "from src.area_utils import compute_area_estimate, compute_area_error_matrix, compute_std_p_i\n", - "from sklearn.metrics import confusion_matrix" - ], - "id": "LY6Q_QtUgME_" - }, - { - "cell_type": "code", - "execution_count": 87, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "oojPqwSboiWU", - "outputId": "091f0d2b-2c73-42af-8055-e28ca741ef2c" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ], - "source": [ - "def compute_area_estimate(dataset, true, pred, a_j, resolution):\n", - " cm = confusion_matrix(true, pred)\n", - " total_px = a_j.sum()\n", - " w_j = a_j / total_px\n", - "\n", - " am = compute_area_error_matrix(cm, w_j)\n", - " a_i = am.sum(axis=1)\n", - " std_a_i = compute_std_p_i(w_j, am, cm)\n", - " err_a_i = 1.96 * std_a_i\n", - "\n", - " a_px = total_px * a_i\n", - " err_px = err_a_i * total_px\n", - " return pd.DataFrame(\n", - " data={\n", - " \"dataset\": dataset,\n", - " \"area_ha\": a_px[1] * (resolution**2) / (100**2),\n", - " \"err_ha\": err_px[1] * (resolution**2) / (100**2),\n", - " },\n", - " index=[0],\n", - " ).round(2)" - ], - "id": "oojPqwSboiWU" - }, - { - "cell_type": "code", - "execution_count": 111, - "metadata": { - "id": "ti5ZXmbyn6Mm", - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "outputId": "10b229e0-28ff-4690-d571-ce286ac789df" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - } - ], - "source": [ - "comparisons = []\n", - "area_est = []\n", - "for cropmap in TARGETS.values():\n", - " cropmap, resolution = cropmap.title, cropmap.resolution\n", - " if cropmap not in gdf.columns:\n", - " continue\n", - " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", - " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", - " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", - " comparisons.append(comparison)\n", - " area_est.append(area)\n", - "\n", - "# # Add ensemble\n", - "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", - "\n", - "# print(f\"Ensemble maps: {ensemble_maps}\")\n", - "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", - "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", - "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", - "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", - "# comparisons.append(comparison)\n", - "# area_est.append(area)\n", - "\n", - "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", - "area_est = pd.concat(area_est).set_index(['dataset'])\n", - "\n", - "results = comparisons.merge(area_est, on='dataset')" - ], - "id": "ti5ZXmbyn6Mm" - }, - { - "cell_type": "code", - "execution_count": null, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 17 - }, - "id": "QrAgv7pP1lcz", - "outputId": "6f33c955-6ceb-4295-84ed-4aaf65c1512f" - }, - "outputs": [ - { - "data": { - "text/html": [ - "\n", - " \n", - " " - ], - "text/plain": [ - "" - ] - }, - "metadata": {}, - "output_type": "display_data" - } - ], - "source": [ - "results.to_csv('results.csv')" - ], - "id": "QrAgv7pP1lcz" - }, - { - "cell_type": "code", - "source": [ - "results.columns" - ], - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 124 - }, - "id": "xOO6fdt0CiG6", - "outputId": "5ba33666-7b75-4785-915c-5a2cb3a7d12f" - }, - "id": "xOO6fdt0CiG6", - "execution_count": null, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "execute_result", - "data": { - "text/plain": [ - "Index(['country', 'crop_f1', 'std_crop_f1', 'accuracy', 'std_acc',\n", - " 'crop_recall_pa', 'std_crop_pa', 'noncrop_recall_pa', 'std_noncrop_pa',\n", - " 'crop_precision_ua', 'std_crop_ua', 'noncrop_precision_ua',\n", - " 'std_noncrop_ua', 'crop_support', 'noncrop_support', 'tn', 'fp', 'fn',\n", - " 'tp', 'tn_area', 'fp_area', 'fn_area', 'tp_area', 'area_ha', 'err_ha'],\n", - " dtype='object')" - ] - }, - "metadata": {}, - "execution_count": 21 - } - ] - }, - { - "cell_type": "code", - "execution_count": 89, - "metadata": { - "colab": { - "base_uri": "https://localhost:8080/", - "height": 175 - }, - "id": "nAj0p7VS1_2K", - "outputId": "e2f2e6fd-542f-4a26-89e7-0cc017cd4753" - }, - "outputs": [ - { - "output_type": "display_data", - "data": { - "text/plain": [ - "" - ], - "text/html": [ - "\n", - " \n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "execute_result", - "data": { - "text/plain": [ - " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", - "dataset \n", - "worldcover-v100 0.64 0.88 0.01 0.66 0.04 \n", - "worldcover-v200 0.65 0.89 0.01 0.69 0.04 \n", - "worldcereal-v100 0.64 0.88 0.01 0.66 0.04 \n", - "\n", - " crop_precision_ua std_crop_ua area_ha err_ha \n", - "dataset \n", - "worldcover-v100 0.61 0.05 3097982.37 486283.35 \n", - "worldcover-v200 0.62 0.05 3068403.67 475012.65 \n", - "worldcereal-v100 0.62 0.05 3184080.79 495468.16 " - ], - "text/html": [ - "\n", - "
\n", - "
\n", - "\n", - "\n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - " \n", - "
crop_f1accuracystd_acccrop_recall_pastd_crop_pacrop_precision_uastd_crop_uaarea_haerr_ha
dataset
worldcover-v1000.640.880.010.660.040.610.053097982.37486283.35
worldcover-v2000.650.890.010.690.040.620.053068403.67475012.65
worldcereal-v1000.640.880.010.660.040.620.053184080.79495468.16
\n", - "
\n", - "
\n", - "\n", - "
\n", - " \n", - "\n", - " \n", - "\n", - " \n", - "
\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "reference_year = 2022\n", + "TARGETS = {k: v for k, v in TARGETS.items() if v.year in range(reference_year - 2, reference_year + 1)}\n", + "# TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 2, reference_year, reference_year + 2]}" + ], + "id": "1oQjubrHjkBi" + }, + { + "cell_type": "code", + "execution_count": 14, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 71 + }, + "id": "98e241d2", + "outputId": "14d937ca-a3fa-49ac-9874-59719991166c" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ "\n", - "\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[Senegal] sampling worldcover-v100...\n", + "[Senegal] sampling worldcover-v200...\n", + "[Senegal] sampling worldcereal-v100...\n" + ] + } + ], + "source": [ + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] sampling \" + cropmap.title + \"...\")\n", + " map_sampled = cropmap.extract_test(gdf).copy()\n", + " gdf = pd.merge(gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", + " gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" + ], + "id": "98e241d2" + }, + { + "cell_type": "code", + "execution_count": 15, + "metadata": { + "id": "95a0f536", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 71 + }, + "outputId": "306aaab2-d317-459f-838c-df727edf0358" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ "\n", - " .colab-df-quickchart-complete:disabled,\n", - " .colab-df-quickchart-complete:disabled:hover {\n", - " background-color: var(--disabled-bg-color);\n", - " fill: var(--disabled-fill-color);\n", - " box-shadow: none;\n", - " }\n", + " \n", + " .geemap-colab {\n", + " background-color: var(--colab-primary-surface-color, white);\n", + " }\n", "\n", - " \n", - "
\n", - "
\n", - "
\n" - ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "summary": "{\n \"name\": \"results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"dataset\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"worldcover-v100\",\n \"worldcover-v200\",\n \"worldcereal-v100\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_f1\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896263,\n \"min\": 0.64,\n \"max\": 0.65,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.65,\n 0.64\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"accuracy\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896262,\n \"min\": 0.88,\n \"max\": 0.89,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.89,\n 0.88\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_acc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.01,\n \"max\": 0.01,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_recall_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.017320508075688724,\n \"min\": 0.66,\n \"max\": 0.69,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.69\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.04,\n \"max\": 0.04,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.04\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_precision_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896262,\n \"min\": 0.61,\n \"max\": 0.62,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.62\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8.498374721940739e-18,\n \"min\": 0.05,\n \"max\": 0.05,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.05\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"area_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 60095.80339158582,\n \"min\": 3068403.67,\n \"max\": 3184080.79,\n \"num_unique_values\": 3,\n \"samples\": [\n 3097982.37\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"err_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 10245.46487139714,\n \"min\": 475012.65,\n \"max\": 495468.16,\n \"num_unique_values\": 3,\n \"samples\": [\n 486283.35\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } + " .geemap-colab .jupyter-button {\n", + " --jp-layout-color3: var(--colab-primary-surface-color, white);\n", + " }\n", + " \n", + " " + ] }, - "metadata": {}, - "execution_count": 89 + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "[Senegal] calculating pixel area for worldcover-v100...\n", + "[Senegal] calculating pixel area for worldcover-v200...\n", + "[Senegal] calculating pixel area for worldcereal-v100...\n" + ] } ], "source": [ - "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI" + "a_j = {}\n", + "for cropmap in TARGETS.values():\n", + " if country not in cropmap.countries:\n", + " continue\n", + " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", + " # a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", + " # a_j[cropmap.title] = cropmap.compute_map_area(country, dataset_name=cropmap.title).copy()\n", + " a_j[cropmap.title] = np.array([None,None])" ], - "id": "nAj0p7VS1_2K" + "id": "95a0f536" }, { - "cell_type": "markdown", - "metadata": { - "id": "fa969373" - }, + "cell_type": "code", "source": [ - "## 4. Visualize best available map" + "# update a_j values with exported values\n", + "for cropmap in a_j.keys():\n", + " try:\n", + " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-{country}-{cropmap}.csv')\n", + " except:\n", + " continue\n", + " crop_area = int(area_df['crop_sum'][0])\n", + " noncrop_area = int(area_df['noncrop_sum'][0])\n", + " a_j[cropmap] = np.array([noncrop_area, crop_area])" ], - "id": "fa969373" + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "5fJPzvOeUo9G", + "outputId": "cd2104b3-fa8c-416b-9e83-6720fc302f94" + }, + "id": "5fJPzvOeUo9G", + "execution_count": 16, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ] }, { "cell_type": "code", "source": [ - "results.dropna(inplace=True)" + "# Change None to nan\n", + "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" ], "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 17 }, - "id": "qenOtnORfGTR", - "outputId": "e4430a49-6836-44be-b753-86f58ac8e387" + "id": "zyR4qCJ49Rh5", + "outputId": "9df5e7f7-7010-4259-db7c-f8c949b6dc59" }, - "id": "qenOtnORfGTR", - "execution_count": 29, + "id": "zyR4qCJ49Rh5", + "execution_count": 17, "outputs": [ { "output_type": "display_data", @@ -3031,14 +1123,14 @@ }, { "cell_type": "code", - "execution_count": 90, + "execution_count": 18, "metadata": { "colab": { "base_uri": "https://localhost:8080/", - "height": 1000 + "height": 17 }, - "id": "fraQjcTMpTwp", - "outputId": "4b4b7209-c8c7-4edc-9c67-3c8145bf61cd" + "id": "LY6Q_QtUgME_", + "outputId": "90faaa32-64fa-461a-8463-1e89892df5eb" }, "outputs": [ { @@ -3074,58 +1166,96 @@ ] }, "metadata": {} + } + ], + "source": [ + "from src.area_utils import compute_area_estimate, compute_area_error_matrix, compute_std_p_i\n", + "from sklearn.metrics import confusion_matrix" + ], + "id": "LY6Q_QtUgME_" + }, + { + "cell_type": "code", + "execution_count": 19, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 }, + "id": "oojPqwSboiWU", + "outputId": "4f059e76-205a-4cac-c8f4-c475e6408bca" + }, + "outputs": [ { "output_type": "display_data", "data": { "text/plain": [ - "
" + "" ], - "image/png": "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\n" + "text/html": [ + "\n", + " \n", + " " + ] }, "metadata": {} } ], "source": [ - "crop_proportion = round(gdf[CLASS_COL].value_counts(normalize=True)[1], 4) * 100\n", - "ax = results.sort_values(\"crop_f1\").plot(\n", - " y=[\"accuracy\", \"crop_recall_pa\", \"crop_precision_ua\", \"crop_f1\"],\n", - " xerr=\"std_crop_f1\",\n", - " kind=\"barh\",\n", - " figsize=(6, 14),\n", - " width=0.8,\n", - " title=f\"{country}: {len(gdf)} points (crop proportion: {crop_proportion}%)\",\n", - ");\n", - "\n", - "for c in ax.containers[1::2]:\n", - " ax.bar_label(c)\n", + "def compute_area_estimate(dataset, true, pred, a_j, resolution):\n", + " cm = confusion_matrix(true, pred)\n", + " total_px = a_j.sum()\n", + " w_j = a_j / total_px\n", "\n", - "for border in [\"top\", \"right\", \"bottom\", \"left\"]:\n", - " ax.spines[border].set_visible(False)\n", + " am = compute_area_error_matrix(cm, w_j)\n", + " a_i = am.sum(axis=1)\n", + " std_a_i = compute_std_p_i(w_j, am, cm)\n", + " err_a_i = 1.96 * std_a_i\n", "\n", - "ax.legend(bbox_to_anchor=(1, 1), reverse=True);" + " a_px = total_px * a_i\n", + " err_px = err_a_i * total_px\n", + " return pd.DataFrame(\n", + " data={\n", + " \"dataset\": dataset,\n", + " \"area_ha\": a_px[1] * (resolution**2) / (100**2),\n", + " \"err_ha\": err_px[1] * (resolution**2) / (100**2),\n", + " },\n", + " index=[0],\n", + " ).round(2)" ], - "id": "fraQjcTMpTwp" + "id": "oojPqwSboiWU" }, { "cell_type": "code", - "source": [ - "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", - "# fao_stat = fao_stat[fao_stat['Area'] == country]\n", - "fao_stat = fao_stat[fao_stat['Area'] == country]['Value'].mean() * 1000\n", - "# fao_stat = fao_stat[fao_stat['Year Code'] == reference_year]['Value'] * 1000\n", - "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" - ], + "execution_count": 30, "metadata": { + "id": "ti5ZXmbyn6Mm", "colab": { "base_uri": "https://localhost:8080/", "height": 17 }, - "id": "L-nrhBekPfcp", - "outputId": "bf29eab8-ff43-4d88-983a-6902728c5eec" + "outputId": "786672a4-f508-49ec-e006-43af6be62a8e" }, - "id": "L-nrhBekPfcp", - "execution_count": 116, "outputs": [ { "output_type": "display_data", @@ -3147,37 +1277,115 @@ " .geemap-dark .jupyter-button {\n", " --jp-layout-color3: #383838;\n", " }\n", - "\n", + "\n", + " .geemap-colab {\n", + " background-color: var(--colab-primary-surface-color, white);\n", + " }\n", + "\n", + " .geemap-colab .jupyter-button {\n", + " --jp-layout-color3: var(--colab-primary-surface-color, white);\n", + " }\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], + "source": [ + "comparisons = []\n", + "area_est = []\n", + "for cropmap in TARGETS.values():\n", + " cropmap, resolution = cropmap.title, cropmap.resolution\n", + " if cropmap not in gdf.columns:\n", + " continue\n", + " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", + " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", + " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", + " comparisons.append(comparison)\n", + " area_est.append(area)\n", + "\n", + "# # Add ensemble\n", + "# ensemble_maps = [\"glad\", \"esri-lulc\"] # Should be odd number\n", + "\n", + "# print(f\"Ensemble maps: {ensemble_maps}\")\n", + "# ensemble = gdf[ensemble_maps].mode(axis='columns')\n", + "# a_j['ensemble-subset'] = get_ensemble_area(country, [TARGETS[name] for name in ensemble_maps])\n", + "# comparison = generate_report(\"ensemble-subset\", country, gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], area_weighted=True)\n", + "# area = compute_area_estimate(\"ensemble-subset\", gdf[CLASS_COL], ensemble, a_j['ensemble-subset'], 10)\n", + "# comparisons.append(comparison)\n", + "# area_est.append(area)\n", + "\n", + "comparisons = pd.concat(comparisons).set_index(['dataset'])\n", + "area_est = pd.concat(area_est).set_index(['dataset'])\n", + "\n", + "results = comparisons.merge(area_est, on='dataset')" + ], + "id": "ti5ZXmbyn6Mm" + }, + { + "cell_type": "code", + "execution_count": null, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "QrAgv7pP1lcz", + "outputId": "6f33c955-6ceb-4295-84ed-4aaf65c1512f" + }, + "outputs": [ + { + "data": { + "text/html": [ + "\n", + " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } - ] + ], + "source": [ + "results.to_csv('results.csv')" + ], + "id": "QrAgv7pP1lcz" }, { "cell_type": "code", - "source": [ - "fao_stat[fao_stat['Area'] == country]" - ], + "execution_count": 22, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 175 }, - "id": "rF9DADLtnX4B", - "outputId": "a5191569-eb17-438c-ac3a-d8ae63c7c449" + "id": "nAj0p7VS1_2K", + "outputId": "d3b3c919-19d6-42c9-85ca-88e1a75d68ba" }, - "id": "rF9DADLtnX4B", - "execution_count": 115, "outputs": [ { "output_type": "display_data", @@ -3217,27 +1425,21 @@ "output_type": "execute_result", "data": { "text/plain": [ - " Domain Code Domain Area Code (M49) Area Element Code Element \\\n", - "56 RL Land Use 686 Senegal 5110 Area \n", - "57 RL Land Use 686 Senegal 5110 Area \n", - "58 RL Land Use 686 Senegal 5110 Area \n", - "59 RL Land Use 686 Senegal 5110 Area \n", - "\n", - " Item Code Item Year Code Year Unit Value Flag \\\n", - "56 6620 Cropland 2018 2018 1000 ha 3758.0 I \n", - "57 6620 Cropland 2019 2019 1000 ha 3795.0 I \n", - "58 6620 Cropland 2020 2020 1000 ha 3830.0 I \n", - "59 6620 Cropland 2021 2021 1000 ha 3911.0 I \n", - "\n", - " Flag Description Note \n", - "56 Imputed value NaN \n", - "57 Imputed value NaN \n", - "58 Imputed value NaN \n", - "59 Imputed value NaN " + " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", + "dataset \n", + "worldcover-v100 0.65 0.89 0.01 0.70 0.03 \n", + "worldcover-v200 0.67 0.90 0.01 0.73 0.03 \n", + "worldcereal-v100 0.64 0.89 0.01 0.68 0.03 \n", + "\n", + " crop_precision_ua std_crop_ua area_ha err_ha \n", + "dataset \n", + "worldcover-v100 0.62 0.04 2991154.22 343812.58 \n", + "worldcover-v200 0.63 0.04 2974111.70 334988.16 \n", + "worldcereal-v100 0.61 0.04 3013651.69 351020.07 " ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", + " " + ] + }, + "metadata": {} + } + ] + }, + { + "cell_type": "code", + "execution_count": 24, + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 1000 + }, + "id": "fraQjcTMpTwp", + "outputId": "590f1091-6821-4b78-fd2e-d49535fd90f4" + }, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "display_data", + "data": { + "text/plain": [ + "
" + ], + "image/png": "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\n" + }, + "metadata": {} + } + ], + "source": [ + "crop_proportion = round(gdf[CLASS_COL].value_counts(normalize=True)[1], 4) * 100\n", + "ax = results.sort_values(\"crop_f1\").plot(\n", + " y=[\"accuracy\", \"crop_recall_pa\", \"crop_precision_ua\", \"crop_f1\"],\n", + " xerr=\"std_crop_f1\",\n", + " kind=\"barh\",\n", + " figsize=(6, 14),\n", + " width=0.8,\n", + " title=f\"{country}: {len(gdf)} points (crop proportion: {crop_proportion}%)\",\n", + ");\n", + "\n", + "for c in ax.containers[1::2]:\n", + " ax.bar_label(c)\n", + "\n", + "for border in [\"top\", \"right\", \"bottom\", \"left\"]:\n", + " ax.spines[border].set_visible(False)\n", + "\n", + "ax.legend(bbox_to_anchor=(1, 1), reverse=True);" + ], + "id": "fraQjcTMpTwp" + }, + { + "cell_type": "code", + "source": [ + "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", + "fao_stat = fao_stat[fao_stat['Area'] == country]['Value'].mean() * 1000 # Using the mean instead, no data for 2022\n", + "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" + ], + "metadata": { + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + }, + "id": "L-nrhBekPfcp", + "outputId": "4456b3b9-3e65-4ade-be82-dabbdae8896e" + }, + "id": "L-nrhBekPfcp", + "execution_count": 32, + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} } ] }, @@ -3603,10 +1973,10 @@ "base_uri": "https://localhost:8080/", "height": 470 }, - "outputId": "0be8683a-47d8-4137-9530-447be6bbd9de" + "outputId": "4d5cb3ef-3f00-4c2f-bfb6-994801f35ebd" }, "id": "a0XEODxnBXW3", - "execution_count": 110, + "execution_count": 33, "outputs": [ { "output_type": "display_data", @@ -3648,7 +2018,7 @@ "text/plain": [ "
" ], - "image/png": "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\n" + "image/png": "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\n" }, "metadata": {} } @@ -3689,4 +2059,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file From 74e2ffd60727161cb6f6af5332af458cf08edac5 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Sat, 23 Mar 2024 23:48:49 +0000 Subject: [PATCH 14/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index 3aa84cb3..bf264002 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -2059,4 +2059,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From 6977c495c3de04260773310aa2f99a01b0bd84e2 Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Sat, 23 Mar 2024 19:50:28 -0400 Subject: [PATCH 15/21] Update --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 576 +++++++++--------- 1 file changed, 288 insertions(+), 288 deletions(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index bf264002..1a0da835 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -2,48 +2,49 @@ "cells": [ { "cell_type": "markdown", + "id": "778ff440", "metadata": { "id": "778ff440" }, "source": [ "# Intercomparison\n", "\n", - "**Author:**\n", + "**Author:** Adebowale Adebayo\n", "\n", - "**Last updated:**\n", + "**Last updated:** March 23, 2024\n", "\n", - "**Description:** Runs intercomparison for [Country Year]\n", + "**Description:** Runs intercomparison and **area estimate** for Senegal 2022\n", "\n", "## 1. Setup" - ], - "id": "778ff440" + ] }, { "cell_type": "code", "execution_count": null, + "id": "fb42d13c", "metadata": { "id": "fb42d13c" }, "outputs": [], "source": [ "# !earthengine authenticate" - ], - "id": "fb42d13c" + ] }, { "cell_type": "code", "execution_count": 1, + "id": "hZ8qzSlB75kl", "metadata": { - "id": "hZ8qzSlB75kl", "colab": { "base_uri": "https://localhost:8080/" }, + "id": "hZ8qzSlB75kl", "outputId": "7ae24c2f-c6e5-4832-8b23-a00a30836ce2" }, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Cloning into 'crop-mask'...\n", "remote: Enumerating objects: 12121, done.\u001b[K\n", @@ -58,12 +59,12 @@ ], "source": [ "!git clone https://github.com/nasaharvest/crop-mask.git" - ], - "id": "hZ8qzSlB75kl" + ] }, { "cell_type": "code", "execution_count": 2, + "id": "1fe-6D3f8LTb", "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -73,8 +74,8 @@ }, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "/content/crop-mask\n" ] @@ -82,14 +83,12 @@ ], "source": [ "%cd crop-mask/" - ], - "id": "1fe-6D3f8LTb" + ] }, { "cell_type": "code", - "source": [ - "!git checkout area-estimate-from-multi-land-cover" - ], + "execution_count": 3, + "id": "V6lTs8Z9Pt-T", "metadata": { "colab": { "base_uri": "https://localhost:8080/" @@ -97,22 +96,24 @@ "id": "V6lTs8Z9Pt-T", "outputId": "9157afbd-b0be-4978-bd00-28bd74e9d17a" }, - "id": "V6lTs8Z9Pt-T", - "execution_count": 3, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Branch 'area-estimate-from-multi-land-cover' set up to track remote branch 'area-estimate-from-multi-land-cover' from 'origin'.\n", "Switched to a new branch 'area-estimate-from-multi-land-cover'\n" ] } + ], + "source": [ + "!git checkout area-estimate-from-multi-land-cover" ] }, { "cell_type": "code", "execution_count": null, + "id": "gEUyxHk9MEU2", "metadata": { "id": "gEUyxHk9MEU2" }, @@ -121,24 +122,24 @@ "!pip install cartopy -qq\n", "!pip install rasterio -qq\n", "!pip install dvc[gs] -qq" - ], - "id": "gEUyxHk9MEU2" + ] }, { "cell_type": "code", "execution_count": 5, + "id": "9907f9a5", "metadata": { - "id": "9907f9a5", "colab": { "base_uri": "https://localhost:8080/", "height": 73 }, + "id": "9907f9a5", "outputId": "762ee4ed-a169-43d5-968e-01e71f287bf3" }, "outputs": [ { - "output_type": "stream", "name": "stderr", + "output_type": "stream", "text": [ "/usr/local/lib/python3.10/dist-packages/cartopy/io/__init__.py:241: DownloadWarning: Downloading: https://naturalearth.s3.amazonaws.com/10m_cultural/ne_10m_admin_1_states_provinces.zip\n", " warnings.warn(f'Downloading: {url}', DownloadWarning)\n" @@ -162,22 +163,22 @@ "\n", "from src.compare_covermaps import TARGETS, filter_by_bounds, generate_report, CLASS_COL, COUNTRY_COL, get_ensemble_area\n", "from src.compare_covermaps import TEST_COUNTRIES, TEST_CODE" - ], - "id": "9907f9a5" + ] }, { "cell_type": "markdown", + "id": "c61ea4f8", "metadata": { "id": "c61ea4f8" }, "source": [ "## 2. Read in evaluation set" - ], - "id": "c61ea4f8" + ] }, { "cell_type": "code", "execution_count": 6, + "id": "7f75e567", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -188,11 +189,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -234,12 +235,12 @@ "else:\n", " country_code = TEST_CODE[country]\n", " # dataset_path = \"../\" + TEST_COUNTRIES[country]" - ], - "id": "7f75e567" + ] }, { "cell_type": "code", "execution_count": 7, + "id": "prvHkUXTOe7P", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -250,11 +251,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ "# dataset_path = TEST_COUNTRIES[country]\n", "dataset_path = 'data/datasets/Senegal_CEO_2022.csv'" - ], - "id": "prvHkUXTOe7P" + ] }, { "cell_type": "code", "execution_count": 8, + "id": "vbVX8gFd_N3J", "metadata": { - "id": "vbVX8gFd_N3J", - "outputId": "f943aa0f-f6b4-437c-a70f-6eed2c07e76e", "colab": { "base_uri": "https://localhost:8080/" - } + }, + "id": "vbVX8gFd_N3J", + "outputId": "f943aa0f-f6b4-437c-a70f-6eed2c07e76e" }, "outputs": [ { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "Applying changes |52.0 [00:06, 8.27file/s]\n", "\u001b[32mA\u001b[0m data/datasets/\n", @@ -314,25 +315,12 @@ ], "source": [ "!dvc pull data/datasets" - ], - "id": "vbVX8gFd_N3J" + ] }, { "cell_type": "code", - "source": [ - "if not Path(dataset_path).exists():\n", - " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", - "else:\n", - " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", - " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", - " # use only test data because validation points used for harvest-dev map\n", - " df = df[(df[\"subset\"] == \"validation\") | (df[\"subset\"] == \"testing\")].copy()\n", - " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", - " df[COUNTRY_COL] = country\n", - "\n", - " gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", - " gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" - ], + "execution_count": 10, + "id": "V8XeT-qci7VG", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -341,15 +329,9 @@ "id": "V8XeT-qci7VG", "outputId": "68dfc81c-1b81-4dfc-f590-f07e32ba9542" }, - "id": "V8XeT-qci7VG", - "execution_count": 10, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "if not Path(dataset_path).exists():\n", + " print(f\"WARNING: Dataset: {dataset_path} not found, run `dvc pull data/datasets from root.\")\n", + "else:\n", + " df = pd.read_csv(dataset_path)[[\"lat\", \"lon\", \"class_probability\", \"subset\"]]\n", + " df = df[(df[\"class_probability\"] != 0.5)].copy()\n", + " # use only test data because validation points used for harvest-dev map\n", + " df = df[(df[\"subset\"] == \"validation\") | (df[\"subset\"] == \"testing\")].copy()\n", + " df[CLASS_COL] = (df[\"class_probability\"] > 0.5).astype(int)\n", + " df[COUNTRY_COL] = country\n", + "\n", + " gdf = gpd.GeoDataFrame(df, geometry=gpd.points_from_xy(df.lon, df.lat), crs=\"epsg:4326\")\n", + " gdf = filter_by_bounds(country_code=country_code, gdf=gdf)" ] }, { "cell_type": "markdown", + "id": "31341d98", "metadata": { "id": "31341d98" }, "source": [ "## 3. Run intercomparison" - ], - "id": "31341d98" + ] }, { "cell_type": "code", "execution_count": 11, + "id": "ImkKe6cEB4aB", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -403,11 +404,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "execute_result", "data": { - "text/plain": [ - " lat lon class_probability subset binary country \\\n", - "0 15.033306 -16.937735 0.000000 testing 0 Senegal \n", - "2 16.192133 -14.772795 0.000000 validation 0 Senegal \n", - "3 15.015340 -13.173794 0.000000 validation 0 Senegal \n", - "4 14.799744 -15.329750 0.000000 testing 0 Senegal \n", - "5 14.260755 -14.656014 0.333333 testing 0 Senegal \n", - "\n", - " geometry \n", - "0 POINT (-16.93773 15.03331) \n", - "2 POINT (-14.77279 16.19213) \n", - "3 POINT (-13.17379 15.01534) \n", - "4 POINT (-15.32975 14.79974) \n", - "5 POINT (-14.65601 14.26076) " - ], + "application/vnd.google.colaboratory.intrinsic+json": { + "summary": "{\n \"name\": \"gdf\",\n \"rows\": 1174,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1419422148869118,\n \"min\": 12.33836043,\n \"max\": 16.64129064,\n \"num_unique_values\": 414,\n \"samples\": [\n 13.33549039,\n 12.76955176,\n 12.84141699\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.398664516536919,\n \"min\": -17.17129666,\n \"max\": -11.39512938,\n \"num_unique_values\": 527,\n \"samples\": [\n -14.16194045,\n -14.35058666,\n -15.29381771\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.2867099180404072,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 4,\n \"samples\": [\n 0.3333333333333333,\n 0.6666666666666666,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2,\n \"samples\": [\n \"validation\",\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Senegal\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 1174,\n \"samples\": [\n \"POINT (-16.52450965 14.92550845)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}", + "type": "dataframe", + "variable_name": "gdf" + }, "text/html": [ "\n", "
\n", @@ -747,24 +737,35 @@ "
\n", "
\n" ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "variable_name": "gdf", - "summary": "{\n \"name\": \"gdf\",\n \"rows\": 1174,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1419422148869118,\n \"min\": 12.33836043,\n \"max\": 16.64129064,\n \"num_unique_values\": 414,\n \"samples\": [\n 13.33549039,\n 12.76955176,\n 12.84141699\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.398664516536919,\n \"min\": -17.17129666,\n \"max\": -11.39512938,\n \"num_unique_values\": 527,\n \"samples\": [\n -14.16194045,\n -14.35058666,\n -15.29381771\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.2867099180404072,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 4,\n \"samples\": [\n 0.3333333333333333,\n 0.6666666666666666,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2,\n \"samples\": [\n \"validation\",\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Senegal\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 1174,\n \"samples\": [\n \"POINT (-16.52450965 14.92550845)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } + "text/plain": [ + " lat lon class_probability subset binary country \\\n", + "0 15.033306 -16.937735 0.000000 testing 0 Senegal \n", + "2 16.192133 -14.772795 0.000000 validation 0 Senegal \n", + "3 15.015340 -13.173794 0.000000 validation 0 Senegal \n", + "4 14.799744 -15.329750 0.000000 testing 0 Senegal \n", + "5 14.260755 -14.656014 0.333333 testing 0 Senegal \n", + "\n", + " geometry \n", + "0 POINT (-16.93773 15.03331) \n", + "2 POINT (-14.77279 16.19213) \n", + "3 POINT (-13.17379 15.01534) \n", + "4 POINT (-15.32975 14.79974) \n", + "5 POINT (-14.65601 14.26076) " + ] }, + "execution_count": 11, "metadata": {}, - "execution_count": 11 + "output_type": "execute_result" } ], "source": [ "gdf.head()" - ], - "id": "ImkKe6cEB4aB" + ] }, { "cell_type": "code", "execution_count": 12, + "id": "54c4cc0f", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -775,11 +776,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -816,12 +817,12 @@ " continue\n", " if v.year is None:\n", " v.year = v.collection_years[v.countries.index(country)]" - ], - "id": "54c4cc0f" + ] }, { "cell_type": "code", "execution_count": 13, + "id": "1oQjubrHjkBi", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -832,11 +833,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ "reference_year = 2022\n", "TARGETS = {k: v for k, v in TARGETS.items() if v.year in range(reference_year - 2, reference_year + 1)}\n", "# TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 2, reference_year, reference_year + 2]}" - ], - "id": "1oQjubrHjkBi" + ] }, { "cell_type": "code", "execution_count": 14, + "id": "98e241d2", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -886,11 +887,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "[Senegal] sampling worldcover-v100...\n", "[Senegal] sampling worldcover-v200...\n", @@ -937,27 +938,23 @@ " map_sampled = cropmap.extract_test(gdf).copy()\n", " gdf = pd.merge(gdf, map_sampled, on=[\"lat\", \"lon\"], how=\"left\")\n", " gdf.drop_duplicates(inplace=True) # TODO find why points get duplicated" - ], - "id": "98e241d2" + ] }, { "cell_type": "code", "execution_count": 15, + "id": "95a0f536", "metadata": { - "id": "95a0f536", "colab": { "base_uri": "https://localhost:8080/", "height": 71 }, + "id": "95a0f536", "outputId": "306aaab2-d317-459f-838c-df727edf0358" }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "stream", "name": "stdout", + "output_type": "stream", "text": [ "[Senegal] calculating pixel area for worldcover-v100...\n", "[Senegal] calculating pixel area for worldcover-v200...\n", @@ -1005,22 +1006,12 @@ " # a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", " # a_j[cropmap.title] = cropmap.compute_map_area(country, dataset_name=cropmap.title).copy()\n", " a_j[cropmap.title] = np.array([None,None])" - ], - "id": "95a0f536" + ] }, { "cell_type": "code", - "source": [ - "# update a_j values with exported values\n", - "for cropmap in a_j.keys():\n", - " try:\n", - " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-{country}-{cropmap}.csv')\n", - " except:\n", - " continue\n", - " crop_area = int(area_df['crop_sum'][0])\n", - " noncrop_area = int(area_df['noncrop_sum'][0])\n", - " a_j[cropmap] = np.array([noncrop_area, crop_area])" - ], + "execution_count": 16, + "id": "5fJPzvOeUo9G", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1029,15 +1020,9 @@ "id": "5fJPzvOeUo9G", "outputId": "cd2104b3-fa8c-416b-9e83-6720fc302f94" }, - "id": "5fJPzvOeUo9G", - "execution_count": 16, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "# update a_j values with exported values\n", + "for cropmap in a_j.keys():\n", + " try:\n", + " area_df = pd.read_csv(f'./Crop_NonCrop_Area_Sum_Export-{country}-{cropmap}.csv')\n", + " except:\n", + " continue\n", + " crop_area = int(area_df['crop_sum'][0])\n", + " noncrop_area = int(area_df['noncrop_sum'][0])\n", + " a_j[cropmap] = np.array([noncrop_area, crop_area])" ] }, { "cell_type": "code", - "source": [ - "# Change None to nan\n", - "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" - ], + "execution_count": 17, + "id": "zyR4qCJ49Rh5", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1082,15 +1080,9 @@ "id": "zyR4qCJ49Rh5", "outputId": "9df5e7f7-7010-4259-db7c-f8c949b6dc59" }, - "id": "zyR4qCJ49Rh5", - "execution_count": 17, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "# Change None to nan\n", + "a_j = {k: np.array([np.nan, np.nan]) if np.any(v == None) else v for k,v in a_j.items()}" ] }, { "cell_type": "code", "execution_count": 18, + "id": "LY6Q_QtUgME_", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1134,11 +1135,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ "from src.area_utils import compute_area_estimate, compute_area_error_matrix, compute_std_p_i\n", "from sklearn.metrics import confusion_matrix" - ], - "id": "LY6Q_QtUgME_" + ] }, { "cell_type": "code", "execution_count": 19, + "id": "oojPqwSboiWU", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1187,11 +1188,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -1242,27 +1243,23 @@ " },\n", " index=[0],\n", " ).round(2)" - ], - "id": "oojPqwSboiWU" + ] }, { "cell_type": "code", "execution_count": 30, + "id": "ti5ZXmbyn6Mm", "metadata": { - "id": "ti5ZXmbyn6Mm", "colab": { "base_uri": "https://localhost:8080/", "height": 17 }, + "id": "ti5ZXmbyn6Mm", "outputId": "786672a4-f508-49ec-e006-43af6be62a8e" }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -1320,12 +1321,12 @@ "area_est = pd.concat(area_est).set_index(['dataset'])\n", "\n", "results = comparisons.merge(area_est, on='dataset')" - ], - "id": "ti5ZXmbyn6Mm" + ] }, { "cell_type": "code", "execution_count": null, + "id": "QrAgv7pP1lcz", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1372,12 +1373,12 @@ ], "source": [ "results.to_csv('results.csv')" - ], - "id": "QrAgv7pP1lcz" + ] }, { "cell_type": "code", "execution_count": 22, + "id": "nAj0p7VS1_2K", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1388,11 +1389,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "execute_result", "data": { - "text/plain": [ - " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", - "dataset \n", - "worldcover-v100 0.65 0.89 0.01 0.70 0.03 \n", - "worldcover-v200 0.67 0.90 0.01 0.73 0.03 \n", - "worldcereal-v100 0.64 0.89 0.01 0.68 0.03 \n", - "\n", - " crop_precision_ua std_crop_ua area_ha err_ha \n", - "dataset \n", - "worldcover-v100 0.62 0.04 2991154.22 343812.58 \n", - "worldcover-v200 0.63 0.04 2974111.70 334988.16 \n", - "worldcereal-v100 0.61 0.04 3013651.69 351020.07 " - ], + "application/vnd.google.colaboratory.intrinsic+json": { + "summary": "{\n \"name\": \"results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"dataset\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"worldcover-v100\",\n \"worldcover-v200\",\n \"worldcereal-v100\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_f1\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.01527525231651948,\n \"min\": 0.64,\n \"max\": 0.67,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.65,\n 0.67,\n 0.64\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"accuracy\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896263,\n \"min\": 0.89,\n \"max\": 0.9,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.9,\n 0.89\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_acc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.01,\n \"max\": 0.01,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_recall_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.025166114784235805,\n \"min\": 0.68,\n \"max\": 0.73,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.7\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.03,\n \"max\": 0.03,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.03\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_precision_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.010000000000000009,\n \"min\": 0.61,\n \"max\": 0.63,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.62\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.04,\n \"max\": 0.04,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.04\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"area_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 19832.60973561544,\n \"min\": 2974111.7,\n \"max\": 3013651.69,\n \"num_unique_values\": 3,\n \"samples\": [\n 2991154.22\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"err_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8029.533389147442,\n \"min\": 334988.16,\n \"max\": 351020.07,\n \"num_unique_values\": 3,\n \"samples\": [\n 343812.58\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}", + "type": "dataframe" + }, "text/html": [ "\n", "
\n", @@ -1730,35 +1721,43 @@ "
\n", "
\n" ], - "application/vnd.google.colaboratory.intrinsic+json": { - "type": "dataframe", - "summary": "{\n \"name\": \"results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"dataset\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"worldcover-v100\",\n \"worldcover-v200\",\n \"worldcereal-v100\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_f1\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.01527525231651948,\n \"min\": 0.64,\n \"max\": 0.67,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.65,\n 0.67,\n 0.64\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"accuracy\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896263,\n \"min\": 0.89,\n \"max\": 0.9,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.9,\n 0.89\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_acc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.01,\n \"max\": 0.01,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_recall_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.025166114784235805,\n \"min\": 0.68,\n \"max\": 0.73,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.7\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.03,\n \"max\": 0.03,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.03\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_precision_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.010000000000000009,\n \"min\": 0.61,\n \"max\": 0.63,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.62\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.04,\n \"max\": 0.04,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.04\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"area_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 19832.60973561544,\n \"min\": 2974111.7,\n \"max\": 3013651.69,\n \"num_unique_values\": 3,\n \"samples\": [\n 2991154.22\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"err_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8029.533389147442,\n \"min\": 334988.16,\n \"max\": 351020.07,\n \"num_unique_values\": 3,\n \"samples\": [\n 343812.58\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}" - } + "text/plain": [ + " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", + "dataset \n", + "worldcover-v100 0.65 0.89 0.01 0.70 0.03 \n", + "worldcover-v200 0.67 0.90 0.01 0.73 0.03 \n", + "worldcereal-v100 0.64 0.89 0.01 0.68 0.03 \n", + "\n", + " crop_precision_ua std_crop_ua area_ha err_ha \n", + "dataset \n", + "worldcover-v100 0.62 0.04 2991154.22 343812.58 \n", + "worldcover-v200 0.63 0.04 2974111.70 334988.16 \n", + "worldcereal-v100 0.61 0.04 3013651.69 351020.07 " + ] }, + "execution_count": 22, "metadata": {}, - "execution_count": 22 + "output_type": "execute_result" } ], "source": [ "results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI" - ], - "id": "nAj0p7VS1_2K" + ] }, { "cell_type": "markdown", + "id": "fa969373", "metadata": { "id": "fa969373" }, "source": [ "## 4. Visualize best available map" - ], - "id": "fa969373" + ] }, { "cell_type": "code", - "source": [ - "results.dropna(inplace=True)" - ], + "execution_count": 23, + "id": "qenOtnORfGTR", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1767,15 +1766,9 @@ "id": "qenOtnORfGTR", "outputId": "04062cf7-5a9c-4e08-ea82-80c9f04f8151" }, - "id": "qenOtnORfGTR", - "execution_count": 23, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "results.dropna(inplace=True)" ] }, { "cell_type": "code", "execution_count": 24, + "id": "fraQjcTMpTwp", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1819,11 +1820,7 @@ }, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "display_data", "data": { + "image/png": "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", "text/plain": [ "
" - ], - "image/png": "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\n" + ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } ], "source": [ @@ -1881,16 +1882,12 @@ " ax.spines[border].set_visible(False)\n", "\n", "ax.legend(bbox_to_anchor=(1, 1), reverse=True);" - ], - "id": "fraQjcTMpTwp" + ] }, { "cell_type": "code", - "source": [ - "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", - "fao_stat = fao_stat[fao_stat['Area'] == country]['Value'].mean() * 1000 # Using the mean instead, no data for 2022\n", - "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" - ], + "execution_count": 32, + "id": "L-nrhBekPfcp", "metadata": { "colab": { "base_uri": "https://localhost:8080/", @@ -1899,15 +1896,9 @@ "id": "L-nrhBekPfcp", "outputId": "4456b3b9-3e65-4ade-be82-dabbdae8896e" }, - "id": "L-nrhBekPfcp", - "execution_count": 32, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", + "fao_stat = fao_stat[fao_stat['Area'] == country]['Value'].mean() * 1000 # Using the mean instead, no data for 2022\n", + "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" ] }, { "cell_type": "code", - "source": [ - "import numpy as np\n", - "import matplotlib.pyplot as plt\n", - "\n", - "import matplotlib.pyplot as plt\n", - "fig, ax = plt.subplots()\n", - "\n", - "n = len(results)\n", - "colors = plt.cm.viridis(np.linspace(0, 1, n))\n", - "\n", - "ax.barh(\n", - " results.index,\n", - " results[\"area_ha\"],\n", - " xerr=results[\"err_ha\"],\n", - " align=\"center\",\n", - " alpha=0.5,\n", - " ecolor=\"black\",\n", - " color= colors\n", - ")\n", - "\n", - "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", - " ax.text(value, i, f\"{value:,} ± {err:,}\", ha=\"center\", va=\"bottom\")\n", - "ax.set_ylabel(\"Area (ha)\")\n", - "ax.set_title(\"Area of cropland\")\n", - "ax.spines[\"right\"].set_visible(False)\n", - "plt.show()" - ], + "execution_count": 33, + "id": "a0XEODxnBXW3", "metadata": { - "id": "a0XEODxnBXW3", "colab": { "base_uri": "https://localhost:8080/", "height": 470 }, + "id": "a0XEODxnBXW3", "outputId": "4d5cb3ef-3f00-4c2f-bfb6-994801f35ebd" }, - "id": "a0XEODxnBXW3", - "execution_count": 33, "outputs": [ { - "output_type": "display_data", "data": { - "text/plain": [ - "" - ], "text/html": [ "\n", " \n", " " + ], + "text/plain": [ + "" ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" }, { - "output_type": "display_data", "data": { + "image/png": "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", "text/plain": [ "
" - ], - "image/png": "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\n" + ] }, - "metadata": {} + "metadata": {}, + "output_type": "display_data" } + ], + "source": [ + "import numpy as np\n", + "import matplotlib.pyplot as plt\n", + "\n", + "import matplotlib.pyplot as plt\n", + "fig, ax = plt.subplots()\n", + "\n", + "n = len(results)\n", + "colors = plt.cm.viridis(np.linspace(0, 1, n))\n", + "\n", + "ax.barh(\n", + " results.index,\n", + " results[\"area_ha\"],\n", + " xerr=results[\"err_ha\"],\n", + " align=\"center\",\n", + " alpha=0.5,\n", + " ecolor=\"black\",\n", + " color= colors\n", + ")\n", + "\n", + "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", + " ax.text(value, i, f\"{value:,} ± {err:,}\", ha=\"center\", va=\"bottom\")\n", + "ax.set_ylabel(\"Area (ha)\")\n", + "ax.set_title(\"Area of cropland\")\n", + "ax.spines[\"right\"].set_visible(False)\n", + "plt.show()" ] }, { "cell_type": "code", - "source": [], + "execution_count": null, + "id": "gfTIBg6cwwAZ", "metadata": { "id": "gfTIBg6cwwAZ" }, - "id": "gfTIBg6cwwAZ", - "execution_count": null, - "outputs": [] + "outputs": [], + "source": [] } ], "metadata": { From cf4247dcb2911dd5a75cd9027660db5143160c58 Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Tue, 26 Mar 2024 10:58:51 -0400 Subject: [PATCH 16/21] Add year --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 490 ++++++++---------- 1 file changed, 226 insertions(+), 264 deletions(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index 1a0da835..fc4cb07d 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -32,31 +32,12 @@ }, { "cell_type": "code", - "execution_count": 1, + "execution_count": null, "id": "hZ8qzSlB75kl", "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "hZ8qzSlB75kl", - "outputId": "7ae24c2f-c6e5-4832-8b23-a00a30836ce2" + "id": "hZ8qzSlB75kl" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Cloning into 'crop-mask'...\n", - "remote: Enumerating objects: 12121, done.\u001b[K\n", - "remote: Counting objects: 100% (1485/1485), done.\u001b[K\n", - "remote: Compressing objects: 100% (472/472), done.\u001b[K\n", - "remote: Total 12121 (delta 1100), reused 1215 (delta 986), pack-reused 10636\u001b[K\n", - "Receiving objects: 100% (12121/12121), 125.63 MiB | 8.99 MiB/s, done.\n", - "Resolving deltas: 100% (7861/7861), done.\n", - "Updating files: 100% (208/208), done.\n" - ] - } - ], + "outputs": [], "source": [ "!git clone https://github.com/nasaharvest/crop-mask.git" ] @@ -70,12 +51,12 @@ "base_uri": "https://localhost:8080/" }, "id": "1fe-6D3f8LTb", - "outputId": "95bcc387-eb42-4921-afb5-193377e147af" + "outputId": "38310d10-f29a-4986-bb01-dfea350da433" }, "outputs": [ { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "/content/crop-mask\n" ] @@ -87,25 +68,12 @@ }, { "cell_type": "code", - "execution_count": 3, + "execution_count": null, "id": "V6lTs8Z9Pt-T", "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "V6lTs8Z9Pt-T", - "outputId": "9157afbd-b0be-4978-bd00-28bd74e9d17a" + "id": "V6lTs8Z9Pt-T" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Branch 'area-estimate-from-multi-land-cover' set up to track remote branch 'area-estimate-from-multi-land-cover' from 'origin'.\n", - "Switched to a new branch 'area-estimate-from-multi-land-cover'\n" - ] - } - ], + "outputs": [], "source": [ "!git checkout area-estimate-from-multi-land-cover" ] @@ -134,12 +102,12 @@ "height": 73 }, "id": "9907f9a5", - "outputId": "762ee4ed-a169-43d5-968e-01e71f287bf3" + "outputId": "4f26d272-1c14-4341-b07f-aaf368a7f452" }, "outputs": [ { - "name": "stderr", "output_type": "stream", + "name": "stderr", "text": [ "/usr/local/lib/python3.10/dist-packages/cartopy/io/__init__.py:241: DownloadWarning: Downloading: https://naturalearth.s3.amazonaws.com/10m_cultural/ne_10m_admin_1_states_provinces.zip\n", " warnings.warn(f'Downloading: {url}', DownloadWarning)\n" @@ -185,11 +153,15 @@ "height": 17 }, "id": "7f75e567", - "outputId": "09fdc14d-3af8-4002-f7fb-daf01e212d62" + "outputId": "4894ed37-6780-4a7b-8faa-0984df98fb4b" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -247,11 +215,15 @@ "height": 17 }, "id": "prvHkUXTOe7P", - "outputId": "3fc18281-33d5-4d2e-b371-5b90d054f990" + "outputId": "2bcc021f-673d-4178-be56-1a9608dad321" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -292,34 +260,19 @@ }, { "cell_type": "code", - "execution_count": 8, + "execution_count": null, "id": "vbVX8gFd_N3J", "metadata": { - "colab": { - "base_uri": "https://localhost:8080/" - }, - "id": "vbVX8gFd_N3J", - "outputId": "f943aa0f-f6b4-437c-a70f-6eed2c07e76e" + "id": "vbVX8gFd_N3J" }, - "outputs": [ - { - "name": "stdout", - "output_type": "stream", - "text": [ - "Applying changes |52.0 [00:06, 8.27file/s]\n", - "\u001b[32mA\u001b[0m data/datasets/\n", - "1 file added and 53 files fetched\n", - "\u001b[0m" - ] - } - ], + "outputs": [], "source": [ "!dvc pull data/datasets" ] }, { "cell_type": "code", - "execution_count": 10, + "execution_count": 68, "id": "V8XeT-qci7VG", "metadata": { "colab": { @@ -327,11 +280,15 @@ "height": 17 }, "id": "V8XeT-qci7VG", - "outputId": "68dfc81c-1b81-4dfc-f590-f07e32ba9542" + "outputId": "c109753f-d46d-465d-8080-295c1783661e" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -392,7 +345,7 @@ }, { "cell_type": "code", - "execution_count": 11, + "execution_count": 69, "id": "ImkKe6cEB4aB", "metadata": { "colab": { @@ -400,11 +353,15 @@ "height": 206 }, "id": "ImkKe6cEB4aB", - "outputId": "f0df68d8-dec2-485b-80a7-c1d033d66fee" + "outputId": "b944d6a8-e92c-4e2d-9fad-6d4ee5d257f6" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} }, { + "output_type": "execute_result", "data": { - "application/vnd.google.colaboratory.intrinsic+json": { - "summary": "{\n \"name\": \"gdf\",\n \"rows\": 1174,\n \"fields\": [\n {\n \"column\": \"lat\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.1419422148869118,\n \"min\": 12.33836043,\n \"max\": 16.64129064,\n \"num_unique_values\": 414,\n \"samples\": [\n 13.33549039,\n 12.76955176,\n 12.84141699\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"lon\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 1.398664516536919,\n \"min\": -17.17129666,\n \"max\": -11.39512938,\n \"num_unique_values\": 527,\n \"samples\": [\n -14.16194045,\n -14.35058666,\n -15.29381771\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"class_probability\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.2867099180404072,\n \"min\": 0.0,\n \"max\": 1.0,\n \"num_unique_values\": 4,\n \"samples\": [\n 0.3333333333333333,\n 0.6666666666666666,\n 0.0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"subset\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 2,\n \"samples\": [\n \"validation\",\n \"testing\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"binary\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0,\n \"min\": 0,\n \"max\": 1,\n \"num_unique_values\": 2,\n \"samples\": [\n 1,\n 0\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"country\",\n \"properties\": {\n \"dtype\": \"category\",\n \"num_unique_values\": 1,\n \"samples\": [\n \"Senegal\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"geometry\",\n \"properties\": {\n \"dtype\": \"geometry\",\n \"num_unique_values\": 1174,\n \"samples\": [\n \"POINT (-16.52450965 14.92550845)\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}", - "type": "dataframe", - "variable_name": "gdf" - }, + "text/plain": [ + " lat lon class_probability subset binary country \\\n", + "0 15.033306 -16.937735 0.000000 testing 0 Senegal \n", + "2 16.192133 -14.772795 0.000000 validation 0 Senegal \n", + "3 15.015340 -13.173794 0.000000 validation 0 Senegal \n", + "4 14.799744 -15.329750 0.000000 testing 0 Senegal \n", + "5 14.260755 -14.656014 0.333333 testing 0 Senegal \n", + "\n", + " geometry \n", + "0 POINT (-16.93773 15.03331) \n", + "2 POINT (-14.77279 16.19213) \n", + "3 POINT (-13.17379 15.01534) \n", + "4 POINT (-15.32975 14.79974) \n", + "5 POINT (-14.65601 14.26076) " + ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -821,7 +774,7 @@ }, { "cell_type": "code", - "execution_count": 13, + "execution_count": 21, "id": "1oQjubrHjkBi", "metadata": { "colab": { @@ -829,11 +782,15 @@ "height": 17 }, "id": "1oQjubrHjkBi", - "outputId": "e3c4325c-bd85-464b-e236-b1ab64fbab3e" + "outputId": "da49f633-e609-4cc5-c946-24d0a28c153d" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ "reference_year = 2022\n", - "TARGETS = {k: v for k, v in TARGETS.items() if v.year in range(reference_year - 2, reference_year + 1)}\n", + "TARGETS = {k: v for k, v in TARGETS.items() if v.year in range(reference_year - 2, reference_year + 2)} # Adjusted the year range to include more map products\n", "# TARGETS = {k: v for k, v in TARGETS.items() if v.year in [reference_year - 2, reference_year, reference_year + 2]}" ] }, { "cell_type": "code", - "execution_count": 14, + "execution_count": 22, "id": "98e241d2", "metadata": { "colab": { @@ -883,11 +836,15 @@ "height": 71 }, "id": "98e241d2", - "outputId": "14d937ca-a3fa-49ac-9874-59719991166c" + "outputId": "613d619e-b08c-4b3a-df2d-4719b35c4fb9" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} }, { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "[Senegal] sampling worldcover-v100...\n", "[Senegal] sampling worldcover-v200...\n", @@ -942,7 +895,7 @@ }, { "cell_type": "code", - "execution_count": 15, + "execution_count": 23, "id": "95a0f536", "metadata": { "colab": { @@ -950,11 +903,15 @@ "height": 71 }, "id": "95a0f536", - "outputId": "306aaab2-d317-459f-838c-df727edf0358" + "outputId": "b2c3ab38-6f4f-4e6a-a838-8b5c3d50326c" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} }, { - "name": "stdout", "output_type": "stream", + "name": "stdout", "text": [ "[Senegal] calculating pixel area for worldcover-v100...\n", "[Senegal] calculating pixel area for worldcover-v200...\n", @@ -1004,13 +957,12 @@ " continue\n", " print(f\"[{country}] calculating pixel area for \" + cropmap.title + \"...\")\n", " # a_j[cropmap.title] = cropmap.compute_map_area(country, export=True, dataset_name=cropmap.title).copy() # I already have the export map areas\n", - " # a_j[cropmap.title] = cropmap.compute_map_area(country, dataset_name=cropmap.title).copy()\n", " a_j[cropmap.title] = np.array([None,None])" ] }, { "cell_type": "code", - "execution_count": 16, + "execution_count": 24, "id": "5fJPzvOeUo9G", "metadata": { "colab": { @@ -1018,11 +970,15 @@ "height": 17 }, "id": "5fJPzvOeUo9G", - "outputId": "cd2104b3-fa8c-416b-9e83-6720fc302f94" + "outputId": "03eeee33-19b5-45fc-8df2-5feb8aade1f2" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -1070,7 +1022,7 @@ }, { "cell_type": "code", - "execution_count": 17, + "execution_count": 25, "id": "zyR4qCJ49Rh5", "metadata": { "colab": { @@ -1078,11 +1030,15 @@ "height": 17 }, "id": "zyR4qCJ49Rh5", - "outputId": "9df5e7f7-7010-4259-db7c-f8c949b6dc59" + "outputId": "e4d11d48-53e2-4a9e-bd99-72efb379d0a5" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -1123,7 +1075,7 @@ }, { "cell_type": "code", - "execution_count": 18, + "execution_count": 26, "id": "LY6Q_QtUgME_", "metadata": { "colab": { @@ -1131,11 +1083,15 @@ "height": 17 }, "id": "LY6Q_QtUgME_", - "outputId": "90faaa32-64fa-461a-8463-1e89892df5eb" + "outputId": "691142c7-c26c-4eb7-e91c-7ac4ce59b000" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -1176,7 +1128,7 @@ }, { "cell_type": "code", - "execution_count": 19, + "execution_count": 44, "id": "oojPqwSboiWU", "metadata": { "colab": { @@ -1184,11 +1136,15 @@ "height": 17 }, "id": "oojPqwSboiWU", - "outputId": "4f059e76-205a-4cac-c8f4-c475e6408bca" + "outputId": "f83877e9-55a7-43d3-ec9f-8e30752f0c69" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ - "def compute_area_estimate(dataset, true, pred, a_j, resolution):\n", + "def compute_area_estimate(dataset, true, pred, a_j, resolution, year):\n", " cm = confusion_matrix(true, pred)\n", " total_px = a_j.sum()\n", " w_j = a_j / total_px\n", @@ -1240,6 +1192,7 @@ " \"dataset\": dataset,\n", " \"area_ha\": a_px[1] * (resolution**2) / (100**2),\n", " \"err_ha\": err_px[1] * (resolution**2) / (100**2),\n", + " \"year\": int(year),\n", " },\n", " index=[0],\n", " ).round(2)" @@ -1247,7 +1200,7 @@ }, { "cell_type": "code", - "execution_count": 30, + "execution_count": 57, "id": "ti5ZXmbyn6Mm", "metadata": { "colab": { @@ -1255,11 +1208,15 @@ "height": 17 }, "id": "ti5ZXmbyn6Mm", - "outputId": "786672a4-f508-49ec-e006-43af6be62a8e" + "outputId": "9a1da855-5908-46bf-c0be-82945f64c788" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ "comparisons = []\n", "area_est = []\n", "for cropmap in TARGETS.values():\n", - " cropmap, resolution = cropmap.title, cropmap.resolution\n", + " cropmap, resolution, year = cropmap.title, cropmap.resolution, cropmap.year\n", " if cropmap not in gdf.columns:\n", " continue\n", " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", - " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", + " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution, year)\n", " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", " comparisons.append(comparison)\n", " area_est.append(area)\n", @@ -1377,7 +1330,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": 58, "id": "nAj0p7VS1_2K", "metadata": { "colab": { @@ -1385,11 +1338,15 @@ "height": 175 }, "id": "nAj0p7VS1_2K", - "outputId": "d3b3c919-19d6-42c9-85ca-88e1a75d68ba" + "outputId": "b3fe8257-7e90-43e7-cecf-095c2d328aa8" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} }, { + "output_type": "execute_result", "data": { - "application/vnd.google.colaboratory.intrinsic+json": { - "summary": "{\n \"name\": \"results[['crop_f1','accuracy','std_acc','crop_recall_pa','std_crop_pa','crop_precision_ua','std_crop_ua','area_ha','err_ha']] # include User and producer's accuracy with their 95% CI\",\n \"rows\": 3,\n \"fields\": [\n {\n \"column\": \"dataset\",\n \"properties\": {\n \"dtype\": \"string\",\n \"num_unique_values\": 3,\n \"samples\": [\n \"worldcover-v100\",\n \"worldcover-v200\",\n \"worldcereal-v100\"\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_f1\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.01527525231651948,\n \"min\": 0.64,\n \"max\": 0.67,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.65,\n 0.67,\n 0.64\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"accuracy\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.005773502691896263,\n \"min\": 0.89,\n \"max\": 0.9,\n \"num_unique_values\": 2,\n \"samples\": [\n 0.9,\n 0.89\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_acc\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.01,\n \"max\": 0.01,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.01\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_recall_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.025166114784235805,\n \"min\": 0.68,\n \"max\": 0.73,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.7\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_pa\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.03,\n \"max\": 0.03,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.03\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"crop_precision_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.010000000000000009,\n \"min\": 0.61,\n \"max\": 0.63,\n \"num_unique_values\": 3,\n \"samples\": [\n 0.62\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"std_crop_ua\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 0.0,\n \"min\": 0.04,\n \"max\": 0.04,\n \"num_unique_values\": 1,\n \"samples\": [\n 0.04\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"area_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 19832.60973561544,\n \"min\": 2974111.7,\n \"max\": 3013651.69,\n \"num_unique_values\": 3,\n \"samples\": [\n 2991154.22\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n },\n {\n \"column\": \"err_ha\",\n \"properties\": {\n \"dtype\": \"number\",\n \"std\": 8029.533389147442,\n \"min\": 334988.16,\n \"max\": 351020.07,\n \"num_unique_values\": 3,\n \"samples\": [\n 343812.58\n ],\n \"semantic_type\": \"\",\n \"description\": \"\"\n }\n }\n ]\n}", - "type": "dataframe" - }, + "text/plain": [ + " crop_f1 accuracy std_acc crop_recall_pa std_crop_pa \\\n", + "dataset \n", + "worldcover-v100 0.65 0.89 0.01 0.70 0.03 \n", + "worldcover-v200 0.67 0.90 0.01 0.73 0.03 \n", + "worldcereal-v100 0.64 0.89 0.01 0.68 0.03 \n", + "\n", + " crop_precision_ua std_crop_ua area_ha err_ha year \n", + "dataset \n", + "worldcover-v100 0.62 0.04 2991154.22 343812.58 2020 \n", + "worldcover-v200 0.63 0.04 2974111.70 334988.16 2021 \n", + "worldcereal-v100 0.61 0.04 3013651.69 351020.07 2021 " + ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -1808,7 +1766,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": 31, "id": "fraQjcTMpTwp", "metadata": { "colab": { @@ -1816,11 +1774,15 @@ "height": 1000 }, "id": "fraQjcTMpTwp", - "outputId": "590f1091-6821-4b78-fd2e-d49535fd90f4" + "outputId": "c0189358-63bf-4109-e863-557d3af39789" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} }, { + "output_type": "display_data", "data": { - "image/png": "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", "text/plain": [ "
" - ] + ], + "image/png": "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\n" }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -1886,7 +1844,7 @@ }, { "cell_type": "code", - "execution_count": 32, + "execution_count": 60, "id": "L-nrhBekPfcp", "metadata": { "colab": { @@ -1894,11 +1852,15 @@ "height": 17 }, "id": "L-nrhBekPfcp", - "outputId": "4456b3b9-3e65-4ade-be82-dabbdae8896e" + "outputId": "0d05922f-af8a-4bf2-f03d-d01caeb45e73" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ - "fao_stat = pd.read_csv(\"FAOSTAT_data_en_3-13-2024.csv\")\n", - "fao_stat = fao_stat[fao_stat['Area'] == country]['Value'].mean() * 1000 # Using the mean instead, no data for 2022\n", - "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0}).set_index(['dataset'])])" + "fao_stat = pd.read_csv(\"./data/ref_samples_area/FAOSTAT_data_en_3-13-2024.csv\")\n", + "fao_stat = fao_stat[fao_stat['Area'] == country]\n", + "fao_stat = fao_stat[fao_stat['Year'] == 2021]['Value'] * 1000 # Using 2021, no data for 2022\n", + "results = pd.concat([results, pd.DataFrame({'dataset':['FAOSTAT'], 'area_ha':fao_stat, 'err_ha':0, 'year':2021}).set_index(['dataset'])])" ] }, { "cell_type": "code", - "execution_count": 33, + "execution_count": 66, "id": "a0XEODxnBXW3", "metadata": { "colab": { @@ -1948,11 +1907,15 @@ "height": 470 }, "id": "a0XEODxnBXW3", - "outputId": "4d5cb3ef-3f00-4c2f-bfb6-994801f35ebd" + "outputId": "9ff8bb6d-6315-45be-f1bb-6bda0e61ff9d" }, "outputs": [ { + "output_type": "display_data", "data": { + "text/plain": [ + "" + ], "text/html": [ "\n", " \n", " " - ], - "text/plain": [ - "" ] }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} }, { + "output_type": "display_data", "data": { - "image/png": "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", "text/plain": [ "
" - ] + ], + "image/png": "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\n" }, - "metadata": {}, - "output_type": "display_data" + "metadata": {} } ], "source": [ @@ -2016,6 +1975,9 @@ " color= colors\n", ")\n", "\n", + "ax.set_xticks(ax.get_xticks()); ax.set_yticks(ax.get_yticks())\n", + "ax.set_yticklabels([f\"{dataset} ({int(results.year[dataset])})\" for dataset in results.index])\n", + "\n", "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", " ax.text(value, i, f\"{value:,} ± {err:,}\", ha=\"center\", va=\"bottom\")\n", "ax.set_ylabel(\"Area (ha)\")\n", @@ -2026,13 +1988,13 @@ }, { "cell_type": "code", - "execution_count": null, - "id": "gfTIBg6cwwAZ", + "source": [], "metadata": { - "id": "gfTIBg6cwwAZ" + "id": "vedCX18xNNR-" }, - "outputs": [], - "source": [] + "id": "vedCX18xNNR-", + "execution_count": null, + "outputs": [] } ], "metadata": { @@ -2059,4 +2021,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file From 002121aded0fc35d9dd62ab54a4c4ec057f8f338 Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Tue, 26 Mar 2024 15:01:20 +0000 Subject: [PATCH 17/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index fc4cb07d..13bbde6e 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -2021,4 +2021,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +} From 4a548c1c7f493b640df29af0fcff4140d36459a2 Mon Sep 17 00:00:00 2001 From: Adebowale Daniel Date: Tue, 26 Mar 2024 11:38:06 -0400 Subject: [PATCH 18/21] Update Kenya --- maps/Kenya_2019/Kenya_area_estimate.ipynb | 346 ++++++++++++---------- 1 file changed, 189 insertions(+), 157 deletions(-) diff --git a/maps/Kenya_2019/Kenya_area_estimate.ipynb b/maps/Kenya_2019/Kenya_area_estimate.ipynb index a2e04969..768b5e82 100644 --- a/maps/Kenya_2019/Kenya_area_estimate.ipynb +++ b/maps/Kenya_2019/Kenya_area_estimate.ipynb @@ -34,28 +34,9 @@ "cell_type": "code", "execution_count": null, "metadata": { - "id": "hZ8qzSlB75kl", - "colab": { - "base_uri": "https://localhost:8080/" - }, - "outputId": "80951fe8-fd11-4fd3-ce7a-f92012496b2d" + "id": "hZ8qzSlB75kl" }, - "outputs": [ - { - "output_type": "stream", - "name": "stdout", - "text": [ - "Cloning into 'crop-mask'...\n", - "remote: Enumerating objects: 12074, done.\u001b[K\n", - "remote: Counting objects: 100% (1485/1485), done.\u001b[K\n", - "remote: Compressing objects: 100% (449/449), done.\u001b[K\n", - "remote: Total 12074 (delta 1102), reused 1232 (delta 1009), pack-reused 10589\u001b[K\n", - "Receiving objects: 100% (12074/12074), 125.43 MiB | 11.56 MiB/s, done.\n", - "Resolving deltas: 100% (7824/7824), done.\n", - "Updating files: 100% (208/208), done.\n" - ] - } - ], + "outputs": [], "source": [ "!git clone https://github.com/nasaharvest/crop-mask.git" ], @@ -63,15 +44,50 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 12, "metadata": { "colab": { - "base_uri": "https://localhost:8080/" + "base_uri": "https://localhost:8080/", + "height": 35 }, "id": "1fe-6D3f8LTb", - "outputId": "6c6848be-2e5f-4c10-ce9c-b4dc071a2795" + "outputId": "66566b67-ed93-4144-a4a1-25142c893d4e" }, "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, { "output_type": "stream", "name": "stdout", @@ -101,16 +117,62 @@ }, { "cell_type": "code", + "source": [ + "!git checkout area-estimate-from-multi-land-cover" + ], + "metadata": { + "id": "V-RT5I2kXJt4" + }, + "id": "V-RT5I2kXJt4", "execution_count": null, + "outputs": [] + }, + { + "cell_type": "code", + "execution_count": 6, "metadata": { "id": "9907f9a5", "colab": { "base_uri": "https://localhost:8080/", "height": 73 }, - "outputId": "036bd16e-89fa-496c-84de-d90a68f3f323" + "outputId": "a5344002-4c63-4f07-860c-179f995d958c" }, "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, { "output_type": "stream", "name": "stderr", @@ -152,14 +214,14 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 7, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 17 }, "id": "7f75e567", - "outputId": "406be616-5fe4-4154-f113-b877ea964c93" + "outputId": "4eeacc2e-61f1-41a8-bbef-f082001aa56d" }, "outputs": [ { @@ -214,8 +276,8 @@ { "cell_type": "code", "source": [ - "ceo_set1 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv'\n", - "ceo_set2 = './data/ref_sample/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv'" + "ceo_set1 = './data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-1-sample-data-2024-03-14.csv'\n", + "ceo_set2 = './data/ref_samples_area/ceo-Kenya-Crop-Area-Estimation-Reference-Sample-2019---Set-2-sample-data-2024-03-14.csv'" ], "metadata": { "colab": { @@ -223,10 +285,10 @@ "height": 17 }, "id": "66-YJBNxYAdF", - "outputId": "221a20ee-9808-4181-8354-d1877c544aca" + "outputId": "ad46f2f6-7de6-402c-99e1-9bde45648b34" }, "id": "66-YJBNxYAdF", - "execution_count": null, + "execution_count": 13, "outputs": [ { "output_type": "display_data", @@ -347,10 +409,10 @@ "height": 17 }, "id": "24QIyHfcZOeG", - "outputId": "17236484-162f-45db-a50f-e205d615f46b" + "outputId": "94a3b69d-9f02-4dd3-df91-cc6618ec3cfa" }, "id": "24QIyHfcZOeG", - "execution_count": null, + "execution_count": 14, "outputs": [ { "output_type": "display_data", @@ -400,10 +462,10 @@ "height": 88 }, "id": "QXMdHSHVauqV", - "outputId": "a003c729-6d8f-47d8-827d-ad62206c680b" + "outputId": "bf082047-7270-4f61-c914-61ab732171ab" }, "id": "QXMdHSHVauqV", - "execution_count": null, + "execution_count": 15, "outputs": [ { "output_type": "display_data", @@ -463,14 +525,14 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 16, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 206 }, "id": "ImkKe6cEB4aB", - "outputId": "719752af-c112-4709-e85f-f577a50d0bab" + "outputId": "c3865269-5698-41bb-f3a7-8ee6927df9a0" }, "outputs": [ { @@ -520,7 +582,7 @@ ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", - " " - ] - }, - "metadata": {} - }, - { - "output_type": "stream", - "name": "stderr", - "text": [ - "/content/crop-mask/src/area_utils.py:385: RuntimeWarning: invalid value encountered in divide\n", - " u_j = p_jjs / p_dotjs\n", - "/content/crop-mask/src/area_utils.py:426: RuntimeWarning: invalid value encountered in divide\n", - " return p_iis / p_idots\n", - "/content/crop-mask/src/area_utils.py:454: RuntimeWarning: invalid value encountered in cast\n", - " n_i_px = ((a_j / cm.sum(axis=0) * cm).sum(axis=1)).astype(np.uint64)\n", - "/content/crop-mask/src/area_utils.py:457: RuntimeWarning: invalid value encountered in cast\n", - " n_j_px = a_j.astype(np.uint64)\n", - "/content/crop-mask/src/area_utils.py:476: RuntimeWarning: divide by zero encountered in divide\n", - " expr_3 = 1 / n_i_px**2\n" - ] - } - ], + "outputs": [], "source": [ "comparisons = []\n", "area_est = []\n", "for cropmap in TARGETS.values():\n", - " cropmap, resolution = cropmap.title, cropmap.resolution\n", + " cropmap, resolution, year = cropmap.title, cropmap.resolution, cropmap.year\n", " if cropmap not in gdf.columns:\n", " continue\n", " temp = gdf[[CLASS_COL, cropmap]].dropna()\n", - " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution)\n", + " area = compute_area_estimate(cropmap, temp[CLASS_COL], temp[cropmap], a_j[cropmap], resolution, year)\n", " comparison = generate_report(cropmap, country, temp[CLASS_COL], temp[cropmap], a_j[cropmap], area_weighted=True)\n", " comparisons.append(comparison)\n", " area_est.append(area)\n", @@ -1439,14 +1448,14 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 27, "metadata": { "colab": { "base_uri": "https://localhost:8080/", "height": 300 }, "id": "nAj0p7VS1_2K", - "outputId": "b82ac7d2-451b-46e9-b76c-299f20a0178c" + "outputId": "24843685-4f51-4e78-aaf5-1009f0413432" }, "outputs": [ { @@ -1497,19 +1506,29 @@ "esri-lulc 0.55 0.89 0.01 0.42 0.03 \n", "harvest-crop-maps 0.47 0.92 0.01 0.47 0.05 \n", "\n", - " crop_precision_ua std_crop_ua area_ha err_ha \n", - "dataset \n", - "copernicus 0.54 0.04 8775032.13 1674917.82 \n", - "worldcover-v100 0.91 0.04 11181466.37 1964935.17 \n", - "glad 0.69 0.04 8504157.39 1587943.81 \n", - "dynamicworld 0.00 0.00 NaN NaN \n", - "digital-earth-africa 0.56 0.04 8104337.54 1585223.55 \n", - "esri-lulc 0.76 0.04 8957165.29 1635602.01 \n", - "harvest-crop-maps 0.47 0.03 4536249.04 1366958.88 " + " crop_precision_ua std_crop_ua area_ha err_ha \\\n", + "dataset \n", + "copernicus 0.54 0.04 8775032.13 1674917.82 \n", + "worldcover-v100 0.91 0.04 11181466.37 1964935.17 \n", + "glad 0.69 0.04 8504157.39 1587943.81 \n", + "dynamicworld 0.00 0.00 NaN NaN \n", + "digital-earth-africa 0.56 0.04 8104337.54 1585223.55 \n", + "esri-lulc 0.76 0.04 8957165.29 1635602.01 \n", + "harvest-crop-maps 0.47 0.03 4536249.04 1366958.88 \n", + "\n", + " year \n", + "dataset \n", + "copernicus 2019 \n", + "worldcover-v100 2020 \n", + "glad 2019 \n", + "dynamicworld 2019 \n", + "digital-earth-africa 2019 \n", + "esri-lulc 2019 \n", + "harvest-crop-maps 2019 " ], "text/html": [ "\n", - "
\n", + "
\n", "
\n", "\n", + " " + ] + }, + "metadata": {} + }, + { + "output_type": "stream", + "name": "stdout", + "text": [ + "Cloning into 'crop-mask'...\n", + "remote: Enumerating objects: 12171, done.\u001b[K\n", + "remote: Counting objects: 100% (1603/1603), done.\u001b[K\n", + "remote: Compressing objects: 100% (441/441), done.\u001b[K\n", + "remote: Total 12171 (delta 1209), reused 1380 (delta 1135), pack-reused 10568\u001b[K\n", + "Receiving objects: 100% (12171/12171), 125.89 MiB | 11.33 MiB/s, done.\n", + "Resolving deltas: 100% (7902/7902), done.\n", + "Updating files: 100% (212/212), done.\n" + ] + } + ], "source": [ "!git clone https://github.com/nasaharvest/crop-mask.git" ] }, { "cell_type": "code", - "execution_count": 2, + "execution_count": 71, "id": "1fe-6D3f8LTb", "metadata": { "colab": { - "base_uri": "https://localhost:8080/" + "base_uri": "https://localhost:8080/", + "height": 35 }, "id": "1fe-6D3f8LTb", - "outputId": "38310d10-f29a-4986-bb01-dfea350da433" + "outputId": "39e61b33-f367-42a0-e115-1e4e61d322ab" }, "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + }, { "output_type": "stream", "name": "stdout", "text": [ - "/content/crop-mask\n" + "/content/crop-mask/crop-mask\n" ] } ], @@ -80,12 +169,52 @@ }, { "cell_type": "code", - "execution_count": null, + "execution_count": 73, "id": "gEUyxHk9MEU2", "metadata": { - "id": "gEUyxHk9MEU2" + "id": "gEUyxHk9MEU2", + "outputId": "5d4a0b49-6f3e-452b-8865-139a7775b11e", + "colab": { + "base_uri": "https://localhost:8080/", + "height": 17 + } }, - "outputs": [], + "outputs": [ + { + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} + } + ], "source": [ "!pip install cartopy -qq\n", "!pip install rasterio -qq\n", @@ -94,24 +223,50 @@ }, { "cell_type": "code", - "execution_count": 5, + "execution_count": 74, "id": "9907f9a5", "metadata": { "colab": { "base_uri": "https://localhost:8080/", - "height": 73 + "height": 17 }, "id": "9907f9a5", - "outputId": "4f26d272-1c14-4341-b07f-aaf368a7f452" + "outputId": "593b90ea-fe30-4b14-9a78-ae61a142e9a9" }, "outputs": [ { - "output_type": "stream", - "name": "stderr", - "text": [ - "/usr/local/lib/python3.10/dist-packages/cartopy/io/__init__.py:241: DownloadWarning: Downloading: https://naturalearth.s3.amazonaws.com/10m_cultural/ne_10m_admin_1_states_provinces.zip\n", - " warnings.warn(f'Downloading: {url}', DownloadWarning)\n" - ] + "output_type": "display_data", + "data": { + "text/plain": [ + "" + ], + "text/html": [ + "\n", + " \n", + " " + ] + }, + "metadata": {} } ], "source": [ @@ -145,7 +300,7 @@ }, { "cell_type": "code", - "execution_count": 6, + "execution_count": 75, "id": "7f75e567", "metadata": { "colab": { @@ -153,7 +308,7 @@ "height": 17 }, "id": "7f75e567", - "outputId": "4894ed37-6780-4a7b-8faa-0984df98fb4b" + "outputId": "2dea5488-5790-4ec2-f2cf-35e17ac6cda8" }, "outputs": [ { @@ -207,7 +362,7 @@ }, { "cell_type": "code", - "execution_count": 7, + "execution_count": null, "id": "prvHkUXTOe7P", "metadata": { "colab": { @@ -272,7 +427,7 @@ }, { "cell_type": "code", - "execution_count": 68, + "execution_count": null, "id": "V8XeT-qci7VG", "metadata": { "colab": { @@ -345,7 +500,7 @@ }, { "cell_type": "code", - "execution_count": 69, + "execution_count": null, "id": "ImkKe6cEB4aB", "metadata": { "colab": { @@ -717,7 +872,7 @@ }, { "cell_type": "code", - "execution_count": 10, + "execution_count": null, "id": "54c4cc0f", "metadata": { "colab": { @@ -774,7 +929,7 @@ }, { "cell_type": "code", - "execution_count": 21, + "execution_count": null, "id": "1oQjubrHjkBi", "metadata": { "colab": { @@ -828,7 +983,7 @@ }, { "cell_type": "code", - "execution_count": 22, + "execution_count": null, "id": "98e241d2", "metadata": { "colab": { @@ -895,7 +1050,7 @@ }, { "cell_type": "code", - "execution_count": 23, + "execution_count": null, "id": "95a0f536", "metadata": { "colab": { @@ -962,7 +1117,7 @@ }, { "cell_type": "code", - "execution_count": 24, + "execution_count": null, "id": "5fJPzvOeUo9G", "metadata": { "colab": { @@ -1022,7 +1177,7 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": null, "id": "zyR4qCJ49Rh5", "metadata": { "colab": { @@ -1075,7 +1230,7 @@ }, { "cell_type": "code", - "execution_count": 26, + "execution_count": null, "id": "LY6Q_QtUgME_", "metadata": { "colab": { @@ -1128,7 +1283,7 @@ }, { "cell_type": "code", - "execution_count": 44, + "execution_count": null, "id": "oojPqwSboiWU", "metadata": { "colab": { @@ -1200,7 +1355,7 @@ }, { "cell_type": "code", - "execution_count": 57, + "execution_count": null, "id": "ti5ZXmbyn6Mm", "metadata": { "colab": { @@ -1330,7 +1485,7 @@ }, { "cell_type": "code", - "execution_count": 58, + "execution_count": null, "id": "nAj0p7VS1_2K", "metadata": { "colab": { @@ -1714,7 +1869,7 @@ }, { "cell_type": "code", - "execution_count": 59, + "execution_count": null, "id": "qenOtnORfGTR", "metadata": { "colab": { @@ -1766,7 +1921,7 @@ }, { "cell_type": "code", - "execution_count": 31, + "execution_count": null, "id": "fraQjcTMpTwp", "metadata": { "colab": { @@ -1844,7 +1999,7 @@ }, { "cell_type": "code", - "execution_count": 60, + "execution_count": null, "id": "L-nrhBekPfcp", "metadata": { "colab": { @@ -1899,15 +2054,15 @@ }, { "cell_type": "code", - "execution_count": 66, + "execution_count": 76, "id": "a0XEODxnBXW3", "metadata": { "colab": { "base_uri": "https://localhost:8080/", - "height": 470 + "height": 472 }, "id": "a0XEODxnBXW3", - "outputId": "9ff8bb6d-6315-45be-f1bb-6bda0e61ff9d" + "outputId": "3224743f-4d61-4603-9441-3ec098bc90c3" }, "outputs": [ { @@ -1950,7 +2105,7 @@ "text/plain": [ "
" ], - "image/png": "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\n" + "image/png": "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\n" }, "metadata": {} } @@ -1980,7 +2135,8 @@ "\n", "for i, (value, err) in enumerate(zip(results[\"area_ha\"], results[\"err_ha\"])):\n", " ax.text(value, i, f\"{value:,} ± {err:,}\", ha=\"center\", va=\"bottom\")\n", - "ax.set_ylabel(\"Area (ha)\")\n", + "ax.set_xlabel(\"Area (ha)\")\n", + "ax.set_ylabel(\"dataset\")\n", "ax.set_title(\"Area of cropland\")\n", "ax.spines[\"right\"].set_visible(False)\n", "plt.show()" @@ -2021,4 +2177,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} +} \ No newline at end of file From b8dc58b8974e07a11dc7f08854ab8873e26df21c Mon Sep 17 00:00:00 2001 From: "pre-commit-ci[bot]" <66853113+pre-commit-ci[bot]@users.noreply.github.com> Date: Tue, 26 Mar 2024 15:42:24 +0000 Subject: [PATCH 21/21] [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci --- maps/Senegal_2022/Senegal_area_estimate.ipynb | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/maps/Senegal_2022/Senegal_area_estimate.ipynb b/maps/Senegal_2022/Senegal_area_estimate.ipynb index f3136540..1996a6bf 100644 --- a/maps/Senegal_2022/Senegal_area_estimate.ipynb +++ b/maps/Senegal_2022/Senegal_area_estimate.ipynb @@ -2177,4 +2177,4 @@ }, "nbformat": 4, "nbformat_minor": 5 -} \ No newline at end of file +}