dataset_stats.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "153a3360",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M1 Pro\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-05-12 01:14:43.399318: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:305] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support.\n",
      "2022-05-12 01:14:43.399443: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:271] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>)\n"
     ]
    }
   ],
   "source": [
    "%load_ext autoreload\n",
    "%autoreload 2\n",
    "\n",
    "\n",
    "from functools import partial\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import IPython\n",
    "import tensorflow as tf\n",
    "\n",
    "import prepare_data\n",
    "import training\n",
    "import utils\n",
    "import models\n",
    "import gin\n",
    "\n",
    "gin.parse_config_file('config/training.gin')\n",
    "gin.enter_interactive_mode()\n",
    "\n",
    "fft_length = gin.query_parameter('%fft_length')\n",
    "frame_length = gin.query_parameter('%frame_length')\n",
    "frame_step = gin.query_parameter('%frame_step')\n",
    "sample_rate = gin.query_parameter('%sample_rate')\n",
    "\n",
    "frames_threshold = sample_rate * 4 / frame_step"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "1c6e49a2",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_dataset, _ = prepare_data.datasets(batch_size=1, prefetch=False, take_batches=None, frames_threshold=None)\n",
    "\n",
    "frame_lengths = []\n",
    "transcription_lengths = []\n",
    "\n",
    "for spectrograms, _mel_spectrograms, transcription in iter(train_dataset):\n",
    "    frame_lengths.append(tf.shape(spectrograms)[1].numpy())\n",
    "    transcription_lengths.append(tf.shape(transcription)[1].numpy())\n",
    "    \n",
    "df = pd.DataFrame({'frames': frame_lengths, 'chars': transcription_lengths})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "08e325c8",
   "metadata": {},
   "outputs": [],
   "source": [
    "df['secs'] = df.frames / (sample_rate * 1.0 / frame_step)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "52a5d7d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>frames</th>\n",
       "      <th>chars</th>\n",
       "      <th>secs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3323</td>\n",
       "      <td>151</td>\n",
       "      <td>13.292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>651</td>\n",
       "      <td>30</td>\n",
       "      <td>2.604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3327</td>\n",
       "      <td>155</td>\n",
       "      <td>13.308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1767</td>\n",
       "      <td>89</td>\n",
       "      <td>7.068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2791</td>\n",
       "      <td>143</td>\n",
       "      <td>11.164</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   frames  chars    secs\n",
       "0    3323    151  13.292\n",
       "1     651     30   2.604\n",
       "2    3327    155  13.308\n",
       "3    1767     89   7.068\n",
       "4    2791    143  11.164"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d0f42731",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.chars.hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "cc2fdf60",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.secs.hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "dc48be4b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<AxesSubplot:>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.frames.hist(bins=100)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "0decc1ef",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x2b4f72250>"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(df.chars, df.secs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e01049f1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1388.000000\n",
       "mean       42.027378\n",
       "std        11.932789\n",
       "min        12.000000\n",
       "25%        34.000000\n",
       "50%        43.000000\n",
       "75%        51.000000\n",
       "max        72.000000\n",
       "Name: chars, dtype: float64"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We might want to start training on small samples first\n",
    "\n",
    "df[df.secs <= 5].chars.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1b130bc4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1739.000000\n",
       "mean        4.151216\n",
       "std         1.177858\n",
       "min         1.516000\n",
       "25%         3.372000\n",
       "50%         4.220000\n",
       "75%         4.972000\n",
       "max        12.108000\n",
       "Name: secs, dtype: float64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Probably we would still like a threshold on the time (number of samples)\n",
    "# To avoid too large batches in the case of outliers\n",
    "\n",
    "df[df.chars <= 60].secs.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "198ae6b7",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}