Inception-ResNet-kPiece-ViT-Model.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "fd1efdf1",
   "metadata": {},
   "source": [
    "### Importing Necessary Libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "0938de48",
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import pickle\n",
    "from tqdm import tqdm\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "import tensorflow as tf\n",
    "from tensorflow.keras.preprocessing.image import load_img, img_to_array\n",
    "\n",
    "from tensorflow import keras\n",
    "from tensorflow.keras.layers import Input, Flatten, Dense, Dropout, GlobalAvgPool2D, GlobalAvgPool1D, BatchNormalization, MultiHeadAttention, Layer, LayerNormalization, Concatenate, MaxPool1D\n",
    "from tensorflow.keras.regularizers import l2\n",
    "\n",
    "from tensorflow.keras.optimizers import Adam\n",
    "\n",
    "from sklearn.metrics import precision_recall_fscore_support as score"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c954b162",
   "metadata": {},
   "source": [
    "## Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "68302673",
   "metadata": {},
   "source": [
    "A detailed discussion and visualization of the data can be seen in [here](Data-Overview.ipynb)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "75b06a70",
   "metadata": {},
   "outputs": [],
   "source": [
    "labels = pd.read_csv(\"./Data/train_labels.csv\")\n",
    "images = ['./Data/train/train/'+str(name)+'.tif' for name in labels['id']]\n",
    "labels['images'] = images\n",
    "labels = labels[['id', 'images', 'label']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "29190fd4",
   "metadata": {},
   "outputs": [],
   "source": [
    "class_labels = list(set(labels['label']))\n",
    "class_names = [\n",
    "    'letter', 'form', 'email', 'handwritten', 'advertisement', 'scientific report', 'scientific publication',\n",
    "    'specification', 'file folder', 'news article', 'budget', 'invoice', 'presentation', 'questionnaire', 'resume',\n",
    "    'memo'\n",
    "]\n",
    "label_names = pd.DataFrame({\n",
    "    'labels': class_labels,\n",
    "    'names': class_names\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "b49810a9",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_labels = pd.read_csv('./Data/train_labels_1.csv')\n",
    "test_labels = pd.read_csv('./Data/test_labels_1.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "ca081f2e",
   "metadata": {},
   "outputs": [],
   "source": [
    "roi_info = np.load('./Data/ROI-Extractions/train/train_roi_viz.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a77be477",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000, 100, 6)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roi_info.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c2e251c1",
   "metadata": {},
   "source": [
    "## Data Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "bb31ae4b",
   "metadata": {},
   "outputs": [],
   "source": [
    "BATCH_SIZE = 50\n",
    "checkpnt_each = 100\n",
    "assert checkpnt_each % BATCH_SIZE == 0\n",
    "checkpnt_i = checkpnt_each // BATCH_SIZE\n",
    "FEATURE_DIM = 1536\n",
    "NUM_ROIS = 100\n",
    "POS_ENC_ANGLE_DENO = 10000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "9fb8348d",
   "metadata": {},
   "outputs": [],
   "source": [
    "roi_count = [np.sum([roi[-1]==1 for roi in doc]) for doc in roi_info] # RoIs per document Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "fbc5e332",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data = list()\n",
    "# for i in range(16000 // (BATCH_SIZE*checkpnt_i)):\n",
    "#     k = 0\n",
    "#     batch_data = np.load(f'./Data/Inception-ResNet-kP-train-precompute/train_data_precomp-{i+1}.npy')\n",
    "#     batch_roi = roi_count[(i*(BATCH_SIZE*checkpnt_i)):((i+1)*(BATCH_SIZE*checkpnt_i))]\n",
    "#     for j in range(BATCH_SIZE*checkpnt_i):\n",
    "#         fine_tune_data.append(np.concatenate([batch_data[k:(k+batch_roi[j])], np.zeros((100-batch_roi[j], FEATURE_DIM))])[np.newaxis,:])\n",
    "#         k += batch_roi[j]\n",
    "# fine_tune_data = np.concatenate(fine_tune_data)\n",
    "# fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "9589ddb3",
   "metadata": {},
   "outputs": [],
   "source": [
    "# np.save(f'./Data/Inception-ResNet-kP-train-precompute/whole_train_data_precomp.npy', fine_tune_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "2f0d6808",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000, 100, 1536)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data = np.load(f'./Data/Inception-ResNet-kP-train-precompute/whole_train_data_precomp.npy')\n",
    "fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "c0cd1c88",
   "metadata": {},
   "outputs": [],
   "source": [
    "mask = np.zeros((*fine_tune_data.shape[:2], fine_tune_data.shape[1]))\n",
    "k = 0\n",
    "for num_roi in roi_count:\n",
    "    mask[k,:num_roi,:num_roi] = 1\n",
    "    k += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "63c7ecb1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000, 512)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fine_tune_data = np.squeeze(np.mean(fine_tune_data, axis=1, keepdims=True))\n",
    "# fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "12bdde76",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000, 7680)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_4P = np.load('./Data/Inception-ResNet-4P-train-precompute/whole_train_data_precomp.npy')\n",
    "fine_tune_data_4P = np.reshape(fine_tune_data_4P, (16000, -1))\n",
    "fine_tune_data_4P.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "6f8bfc7f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data = np.concatenate([fine_tune_data, fine_tune_data_4P], axis=-1)\n",
    "# fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "58c0f4b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# np.save(f'./Data/ResNet-kP-train-precompute/whole_train_data_precomp_roi_n_4p.npy', fine_tune_data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "e1b35850",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data = np.load(f'./Data/ResNet-kP-train-precompute/whole_train_data_precomp_roi_n_4p.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "945aad41",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "568ffcb1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(16000,)"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_labels = labels['label'].to_numpy()\n",
    "fine_tune_labels.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "6e228ab3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15200, 100, 1536)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_train = fine_tune_data[list(train_labels['id'])]\n",
    "fine_tune_data_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "25175484",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15200, 7680)"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_4P_train = fine_tune_data_4P[list(train_labels['id'])]\n",
    "fine_tune_data_4P_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "d42f751e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15200, 100, 100)"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_mask_train = mask[list(train_labels['id'])]\n",
    "fine_tune_mask_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "bcfdf149",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15200,)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_labels_train = train_labels['label'].to_numpy()\n",
    "fine_tune_labels_train.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "ccfac9c8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800, 100, 1536)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_test = fine_tune_data[list(test_labels['id'])]\n",
    "fine_tune_data_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "1db3dc1d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800, 7680)"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_4P_test = fine_tune_data_4P[list(test_labels['id'])]\n",
    "fine_tune_data_4P_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "7815962e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800, 100, 100)"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_mask_test = mask[list(test_labels['id'])]\n",
    "fine_tune_mask_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "132bb5a2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(800,)"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_labels_test = test_labels['label'].to_numpy()\n",
    "fine_tune_labels_test.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48661622",
   "metadata": {},
   "source": [
    "### Validation Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "bc88528c",
   "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>images</th>\n",
       "      <th>id</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>850</th>\n",
       "      <td>./Data/validation/validation/17801.tif</td>\n",
       "      <td>17801</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>898</th>\n",
       "      <td>./Data/validation/validation/17802.tif</td>\n",
       "      <td>17802</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>852</th>\n",
       "      <td>./Data/validation/validation/17803.tif</td>\n",
       "      <td>17803</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>./Data/validation/validation/17804.tif</td>\n",
       "      <td>17804</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>750</th>\n",
       "      <td>./Data/validation/validation/17805.tif</td>\n",
       "      <td>17805</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>./Data/validation/validation/18696.tif</td>\n",
       "      <td>18696</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211</th>\n",
       "      <td>./Data/validation/validation/18697.tif</td>\n",
       "      <td>18697</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>760</th>\n",
       "      <td>./Data/validation/validation/18698.tif</td>\n",
       "      <td>18698</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>787</th>\n",
       "      <td>./Data/validation/validation/18699.tif</td>\n",
       "      <td>18699</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>891</th>\n",
       "      <td>./Data/validation/validation/18700.tif</td>\n",
       "      <td>18700</td>\n",
       "      <td>-1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>900 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     images     id  label\n",
       "850  ./Data/validation/validation/17801.tif  17801     -1\n",
       "898  ./Data/validation/validation/17802.tif  17802     -1\n",
       "852  ./Data/validation/validation/17803.tif  17803     -1\n",
       "798  ./Data/validation/validation/17804.tif  17804     -1\n",
       "750  ./Data/validation/validation/17805.tif  17805     -1\n",
       "..                                      ...    ...    ...\n",
       "246  ./Data/validation/validation/18696.tif  18696     -1\n",
       "211  ./Data/validation/validation/18697.tif  18697     -1\n",
       "760  ./Data/validation/validation/18698.tif  18698     -1\n",
       "787  ./Data/validation/validation/18699.tif  18699     -1\n",
       "891  ./Data/validation/validation/18700.tif  18700     -1\n",
       "\n",
       "[900 rows x 3 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "validation_data = pd.DataFrame({'images':['./Data/validation/validation/'+name for name in os.listdir('./Data/validation/validation/')]})\n",
    "validation_data['id'] = [name.split('.')[0] for name in os.listdir('./Data/validation/validation/')]\n",
    "validation_data['label'] = -1 # Simply added to prevent re-writing code\n",
    "validation_data.sort_values(by=['id'], inplace=True)\n",
    "validation_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "bf4cf1be",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(900, 100, 6)"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "roi_info_validation = np.load('./Data/ROI-Extractions/validation/validation_roi_viz.npy')\n",
    "roi_info_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "91896253",
   "metadata": {},
   "outputs": [],
   "source": [
    "roi_count_validation = [np.sum([roi[-1]==1 for roi in doc]) for doc in roi_info_validation] # RoIs per document Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "cf514e19",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(900, 100, 1536)"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# fine_tune_data_validation = list()\n",
    "# for i in range(900 // (BATCH_SIZE*checkpnt_i)):\n",
    "#     k = 0\n",
    "#     batch_data = np.load(f'./Data/Inception-ResNet-kP-validation-precompute/validation_data_precomp-{i+1}.npy')\n",
    "#     batch_roi = roi_count_validation[(i*(BATCH_SIZE*checkpnt_i)):((i+1)*(BATCH_SIZE*checkpnt_i))]\n",
    "#     for j in range(BATCH_SIZE*checkpnt_i):\n",
    "#         fine_tune_data_validation.append(np.concatenate([batch_data[k:(k+batch_roi[j])], np.zeros((100-batch_roi[j], FEATURE_DIM))])[np.newaxis,:])\n",
    "#         k += batch_roi[j]\n",
    "# fine_tune_data_validation = np.concatenate(fine_tune_data_validation)\n",
    "# fine_tune_data_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "f5be8342",
   "metadata": {},
   "outputs": [],
   "source": [
    "# np.save(f'./Data/Inception-ResNet-kP-validation-precompute/whole_validation_data_precomp.npy', fine_tune_data_validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "38cb6238",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(900, 100, 1536)"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_validation = np.load(f'./Data/Inception-ResNet-kP-validation-precompute/whole_validation_data_precomp.npy')\n",
    "fine_tune_data_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "4f36a021",
   "metadata": {},
   "outputs": [],
   "source": [
    "mask_validation = np.zeros((*fine_tune_data_validation.shape[:2], fine_tune_data_validation.shape[1]))\n",
    "k = 0\n",
    "for num_roi in roi_count_validation:\n",
    "    mask_validation[k,:num_roi,:num_roi] = 1\n",
    "    k += 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "ec86395a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(900, 100, 100)"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5ef3e336",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data_validation = np.squeeze(np.mean(fine_tune_data_validation, axis=1, keepdims=True))\n",
    "# fine_tune_data_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "e6d23ba7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(900, 7680)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fine_tune_data_4P_validation = np.load('./Data/Inception-ResNet-4P-validation-precompute/whole_validation_data_precomp.npy')\n",
    "fine_tune_data_4P_validation = np.reshape(fine_tune_data_4P_validation, (900, -1))\n",
    "fine_tune_data_4P_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "d4b9fc5a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data_validation = np.concatenate([fine_tune_data_validation, fine_tune_data_4P_validation], axis=-1)\n",
    "# fine_tune_data_validation.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "0fd55b80",
   "metadata": {},
   "outputs": [],
   "source": [
    "# np.save(f'./Data/ResNet-kP-validation-precompute/whole_validation_data_precomp_roi_n_4p.npy', fine_tune_data_validation)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "23851219",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data_validation = np.load(f'./Data/ResNet-kP-validation-precompute/whole_validation_data_precomp_roi_n_4p.npy')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "53ed0146",
   "metadata": {},
   "outputs": [],
   "source": [
    "# fine_tune_data_validation.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dff1f776",
   "metadata": {},
   "source": [
    "## Fine-Tuning Models"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7380947f",
   "metadata": {},
   "source": [
    "The fine-tuning part of our model."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "da5b9fd3",
   "metadata": {},
   "outputs": [],
   "source": [
    "def pos_enc(max_len=NUM_ROIS, d_model=FEATURE_DIM):\n",
    "    # returns the positional encoding matrix which needs to be added to the embedding matrix\n",
    "    angles = np.arange(max_len)[:, np.newaxis] / np.power(POS_ENC_ANGLE_DENO, 2*(np.arange(d_model)[np.newaxis, :]//2/np.float32(d_model)))\n",
    "    pos_encode = np.zeros((max_len, d_model))\n",
    "    pos_encode[:, 0::2] = np.sin(angles[:, 0::2])\n",
    "    pos_encode[:, 1::2] = np.cos(angles[:, 1::2])\n",
    "    return tf.cast(pos_encode[np.newaxis, :], dtype=tf.float32)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e6e60894",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Metal device set to: Apple M1 Max\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 15:30:52.113387: 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",
      "2022-10-09 15:30:52.113628: 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"
     ]
    },
    {
     "data": {
      "text/plain": [
       "TensorShape([1, 100, 1536])"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pos_enc().shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "d7eb5baf",
   "metadata": {},
   "outputs": [],
   "source": [
    "class TransformerBlock(Layer):\n",
    "    # Embed Dimension is FEATURE_DIM\n",
    "    def __init__(self, embed_dim, num_heads, ff_dim, rate=0.1):\n",
    "        super().__init__()\n",
    "        self.att = MultiHeadAttention(num_heads=num_heads, key_dim=embed_dim, kernel_regularizer=l2(5e-5))\n",
    "        self.ffn = keras.Sequential(\n",
    "            [Dense(ff_dim, activation=\"relu\"), Dense(embed_dim)]\n",
    "        )\n",
    "        self.layernorm1 = LayerNormalization(epsilon=1e-6)\n",
    "        self.layernorm2 = LayerNormalization(epsilon=1e-6)\n",
    "        self.dropout1 = Dropout(rate)\n",
    "        self.dropout2 = Dropout(rate)\n",
    "\n",
    "    def call(self, inputs, mask, training):\n",
    "        attn_output = self.att(inputs, inputs, inputs, mask)\n",
    "        attn_output = self.dropout1(attn_output, training=training)\n",
    "        out1 = self.layernorm1(inputs + attn_output)\n",
    "        ffn_output = self.ffn(out1)\n",
    "        ffn_output = self.dropout2(ffn_output, training=training)\n",
    "        return self.layernorm2(out1 + ffn_output)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "96a2bdfe",
   "metadata": {},
   "outputs": [],
   "source": [
    "NUM_HEADS = 4\n",
    "FF_DIM = FEATURE_DIM // 4\n",
    "def ftmodel():\n",
    "    inputs = Input(shape=(5*FEATURE_DIM))\n",
    "    x = Flatten()(inputs)\n",
    "    x = BatchNormalization()(x)\n",
    "    \n",
    "    inputs_roi = Input(shape=(100, FEATURE_DIM))\n",
    "    inputs_roi_mask = Input(shape=(100, 100))\n",
    "    y = inputs_roi + pos_enc()\n",
    "    y = MaxPool1D(strides=4, data_format='channels_first')(inputs_roi)\n",
    "    y = TransformerBlock(y.shape[-1], NUM_HEADS, FF_DIM)(y, inputs_roi_mask)\n",
    "    y = GlobalAvgPool1D()(y)\n",
    "    \n",
    "    x = Concatenate()([x, y])\n",
    "    \n",
    "    top_dropout_rate = 0.3\n",
    "    # x = Dropout(top_dropout_rate, name='top_dropout_1')(x)\n",
    "    # x = Dense(1024, name='top_dense_1', kernel_initializer=\"he_normal\", kernel_regularizer=l2(5e-5), activation='relu')(x)\n",
    "    # x = BatchNormalization()(x)\n",
    "    x = Dropout(top_dropout_rate, name='top_dropout_2')(x)\n",
    "    x = Dense(512, name='top_dense_2', kernel_initializer=\"he_normal\", kernel_regularizer=l2(5e-5), activation='relu')(x)\n",
    "    x = BatchNormalization()(x)\n",
    "    outputs = Dense(16, activation='softmax', name='pred', kernel_regularizer=l2(5e-5))(x)\n",
    "\n",
    "    model1 = tf.keras.Model([inputs, inputs_roi, inputs_roi_mask], outputs, name='Inception-ResNet-FineTune')\n",
    "    return model1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "c6574006",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model: \"Inception-ResNet-FineTune\"\n",
      "__________________________________________________________________________________________________\n",
      " Layer (type)                   Output Shape         Param #     Connected to                     \n",
      "==================================================================================================\n",
      " input_2 (InputLayer)           [(None, 100, 1536)]  0           []                               \n",
      "                                                                                                  \n",
      " input_1 (InputLayer)           [(None, 7680)]       0           []                               \n",
      "                                                                                                  \n",
      " max_pooling1d (MaxPooling1D)   (None, 100, 384)     0           ['input_2[0][0]']                \n",
      "                                                                                                  \n",
      " input_3 (InputLayer)           [(None, 100, 100)]   0           []                               \n",
      "                                                                                                  \n",
      " flatten (Flatten)              (None, 7680)         0           ['input_1[0][0]']                \n",
      "                                                                                                  \n",
      " transformer_block (Transformer  (None, 100, 384)    2661504     ['max_pooling1d[0][0]',          \n",
      " Block)                                                           'input_3[0][0]']                \n",
      "                                                                                                  \n",
      " batch_normalization (BatchNorm  (None, 7680)        30720       ['flatten[0][0]']                \n",
      " alization)                                                                                       \n",
      "                                                                                                  \n",
      " global_average_pooling1d (Glob  (None, 384)         0           ['transformer_block[0][0]']      \n",
      " alAveragePooling1D)                                                                              \n",
      "                                                                                                  \n",
      " concatenate (Concatenate)      (None, 8064)         0           ['batch_normalization[0][0]',    \n",
      "                                                                  'global_average_pooling1d[0][0]'\n",
      "                                                                 ]                                \n",
      "                                                                                                  \n",
      " top_dropout_2 (Dropout)        (None, 8064)         0           ['concatenate[0][0]']            \n",
      "                                                                                                  \n",
      " top_dense_2 (Dense)            (None, 512)          4129280     ['top_dropout_2[0][0]']          \n",
      "                                                                                                  \n",
      " batch_normalization_1 (BatchNo  (None, 512)         2048        ['top_dense_2[0][0]']            \n",
      " rmalization)                                                                                     \n",
      "                                                                                                  \n",
      " pred (Dense)                   (None, 16)           8208        ['batch_normalization_1[0][0]']  \n",
      "                                                                                                  \n",
      "==================================================================================================\n",
      "Total params: 6,831,760\n",
      "Trainable params: 6,815,376\n",
      "Non-trainable params: 16,384\n",
      "__________________________________________________________________________________________________\n"
     ]
    }
   ],
   "source": [
    "model1 = ftmodel()\n",
    "model1.summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "2209d28d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<IPython.core.display.Image object>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tf.keras.utils.plot_model(model1, show_shapes=True, expand_nested=True, to_file='./Images/Inception-kPiece-ViT-Model.png')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e56505e2",
   "metadata": {},
   "source": [
    "### Model Training"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "70b7e206",
   "metadata": {},
   "source": [
    "The fine-tuning will be done on the ResNet's $2048D$ representation vectors for $5$ pieces of the image and the averaged ROI representation vectors."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "96524578",
   "metadata": {},
   "outputs": [],
   "source": [
    "model1.compile(loss = 'sparse_categorical_crossentropy', metrics = ['accuracy'], optimizer = Adam(learning_rate = 0.0003))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "20df8897",
   "metadata": {},
   "outputs": [],
   "source": [
    "EPOCHS = 100\n",
    "checkpoint_filepaths = ['./Models/Inception-ResNet-kP-ViT/auto-Inception-ResNet-FT-model-weight']\n",
    "model_checkpoint_callback = [tf.keras.callbacks.ModelCheckpoint(\n",
    "    filepath=checkpoint_filepath,\n",
    "    save_weights_only=True,\n",
    "    monitor = \"loss\",\n",
    "    mode='min',\n",
    "    save_best_only=True) for checkpoint_filepath in checkpoint_filepaths]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "03354278",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 1/100\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 11:52:08.773555: W tensorflow/core/platform/profile_utils/cpu_utils.cc:128] Failed to get CPU frequency: 0 Hz\n",
      "2022-10-09 11:52:09.512866: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - ETA: 0s - loss: 1.8606 - accuracy: 0.5109"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 11:53:32.052487: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "19/19 [==============================] - 86s 4s/step - loss: 1.8606 - accuracy: 0.5109 - val_loss: 2.2738 - val_accuracy: 0.2488\n",
      "Epoch 2/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 1.2070 - accuracy: 0.6649 - val_loss: 2.1622 - val_accuracy: 0.3312\n",
      "Epoch 3/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 1.0232 - accuracy: 0.7148 - val_loss: 2.0381 - val_accuracy: 0.4487\n",
      "Epoch 4/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 0.8968 - accuracy: 0.7521 - val_loss: 1.9821 - val_accuracy: 0.5163\n",
      "Epoch 5/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 0.7926 - accuracy: 0.7841 - val_loss: 1.9463 - val_accuracy: 0.5962\n",
      "Epoch 6/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.7108 - accuracy: 0.8084 - val_loss: 1.9439 - val_accuracy: 0.5712\n",
      "Epoch 7/100\n",
      "19/19 [==============================] - 85s 5s/step - loss: 0.6418 - accuracy: 0.8268 - val_loss: 1.9621 - val_accuracy: 0.5125\n",
      "Epoch 8/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 0.5764 - accuracy: 0.8526 - val_loss: 1.9871 - val_accuracy: 0.4575\n",
      "Epoch 9/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 0.5213 - accuracy: 0.8710 - val_loss: 1.9970 - val_accuracy: 0.4363\n",
      "Epoch 10/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 0.4694 - accuracy: 0.8867 - val_loss: 1.9569 - val_accuracy: 0.4637\n",
      "Epoch 11/100\n",
      "19/19 [==============================] - 85s 4s/step - loss: 0.4351 - accuracy: 0.8980 - val_loss: 1.9244 - val_accuracy: 0.4825\n",
      "Epoch 12/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.3921 - accuracy: 0.9122 - val_loss: 1.8465 - val_accuracy: 0.5263\n",
      "Epoch 13/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.3626 - accuracy: 0.9221 - val_loss: 1.7546 - val_accuracy: 0.5550\n",
      "Epoch 14/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.3349 - accuracy: 0.9321 - val_loss: 1.7194 - val_accuracy: 0.5437\n",
      "Epoch 15/100\n",
      "19/19 [==============================] - 92s 5s/step - loss: 0.3099 - accuracy: 0.9384 - val_loss: 1.6495 - val_accuracy: 0.5537\n",
      "Epoch 16/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.2819 - accuracy: 0.9495 - val_loss: 1.5615 - val_accuracy: 0.5850\n",
      "Epoch 17/100\n",
      "19/19 [==============================] - 105s 6s/step - loss: 0.2667 - accuracy: 0.9522 - val_loss: 1.4198 - val_accuracy: 0.6413\n",
      "Epoch 18/100\n",
      "19/19 [==============================] - 104s 5s/step - loss: 0.2498 - accuracy: 0.9582 - val_loss: 1.3594 - val_accuracy: 0.6662\n",
      "Epoch 19/100\n",
      "19/19 [==============================] - 100s 5s/step - loss: 0.2315 - accuracy: 0.9634 - val_loss: 1.2819 - val_accuracy: 0.6513\n",
      "Epoch 20/100\n",
      "19/19 [==============================] - 98s 5s/step - loss: 0.2132 - accuracy: 0.9684 - val_loss: 1.2251 - val_accuracy: 0.6750\n",
      "Epoch 21/100\n",
      "19/19 [==============================] - 93s 5s/step - loss: 0.2005 - accuracy: 0.9730 - val_loss: 1.1696 - val_accuracy: 0.6962\n",
      "Epoch 22/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.1939 - accuracy: 0.9752 - val_loss: 1.1347 - val_accuracy: 0.6975\n",
      "Epoch 23/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.1823 - accuracy: 0.9773 - val_loss: 1.0513 - val_accuracy: 0.7200\n",
      "Epoch 24/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.1711 - accuracy: 0.9804 - val_loss: 1.0323 - val_accuracy: 0.7088\n",
      "Epoch 25/100\n",
      "19/19 [==============================] - 95s 5s/step - loss: 0.1690 - accuracy: 0.9814 - val_loss: 0.9917 - val_accuracy: 0.7287\n",
      "Epoch 26/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.1569 - accuracy: 0.9824 - val_loss: 0.9867 - val_accuracy: 0.7350\n",
      "Epoch 27/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.1494 - accuracy: 0.9859 - val_loss: 0.9655 - val_accuracy: 0.7300\n",
      "Epoch 28/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.1414 - accuracy: 0.9891 - val_loss: 0.9456 - val_accuracy: 0.7375\n",
      "Epoch 29/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.1344 - accuracy: 0.9891 - val_loss: 0.9486 - val_accuracy: 0.7375\n",
      "Epoch 30/100\n",
      "19/19 [==============================] - 98s 5s/step - loss: 0.1318 - accuracy: 0.9895 - val_loss: 0.9477 - val_accuracy: 0.7262\n",
      "Epoch 31/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.1273 - accuracy: 0.9903 - val_loss: 0.9643 - val_accuracy: 0.7412\n",
      "Epoch 32/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.1218 - accuracy: 0.9914 - val_loss: 0.9358 - val_accuracy: 0.7487\n",
      "Epoch 33/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.1177 - accuracy: 0.9927 - val_loss: 0.9386 - val_accuracy: 0.7450\n",
      "Epoch 34/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.1132 - accuracy: 0.9929 - val_loss: 0.9517 - val_accuracy: 0.7437\n",
      "Epoch 35/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.1097 - accuracy: 0.9938 - val_loss: 0.9815 - val_accuracy: 0.7500\n",
      "Epoch 36/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.1050 - accuracy: 0.9950 - val_loss: 0.9926 - val_accuracy: 0.7375\n",
      "Epoch 37/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.1041 - accuracy: 0.9950 - val_loss: 0.9991 - val_accuracy: 0.7487\n",
      "Epoch 38/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0993 - accuracy: 0.9954 - val_loss: 1.0058 - val_accuracy: 0.7450\n",
      "Epoch 39/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0975 - accuracy: 0.9960 - val_loss: 0.9725 - val_accuracy: 0.7612\n",
      "Epoch 40/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0959 - accuracy: 0.9963 - val_loss: 1.0003 - val_accuracy: 0.7550\n",
      "Epoch 41/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.0938 - accuracy: 0.9963 - val_loss: 0.9692 - val_accuracy: 0.7512\n",
      "Epoch 42/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.0919 - accuracy: 0.9970 - val_loss: 1.0109 - val_accuracy: 0.7400\n",
      "Epoch 43/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0896 - accuracy: 0.9970 - val_loss: 1.0148 - val_accuracy: 0.7437\n",
      "Epoch 44/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0892 - accuracy: 0.9967 - val_loss: 0.9868 - val_accuracy: 0.7537\n",
      "Epoch 45/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0885 - accuracy: 0.9973 - val_loss: 0.9838 - val_accuracy: 0.7437\n",
      "Epoch 46/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0863 - accuracy: 0.9966 - val_loss: 0.9961 - val_accuracy: 0.7462\n",
      "Epoch 47/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0852 - accuracy: 0.9977 - val_loss: 1.0177 - val_accuracy: 0.7550\n",
      "Epoch 48/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0832 - accuracy: 0.9974 - val_loss: 1.0049 - val_accuracy: 0.7562\n",
      "Epoch 49/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0822 - accuracy: 0.9978 - val_loss: 1.0210 - val_accuracy: 0.7512\n",
      "Epoch 50/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0807 - accuracy: 0.9978 - val_loss: 1.0243 - val_accuracy: 0.7500\n",
      "Epoch 51/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.0796 - accuracy: 0.9979 - val_loss: 1.0333 - val_accuracy: 0.7462\n",
      "Epoch 52/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.0802 - accuracy: 0.9974 - val_loss: 1.0026 - val_accuracy: 0.7612\n",
      "Epoch 53/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0779 - accuracy: 0.9976 - val_loss: 0.9881 - val_accuracy: 0.7550\n",
      "Epoch 54/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0763 - accuracy: 0.9986 - val_loss: 1.0124 - val_accuracy: 0.7575\n",
      "Epoch 55/100\n",
      "19/19 [==============================] - 86s 5s/step - loss: 0.0748 - accuracy: 0.9983 - val_loss: 1.0157 - val_accuracy: 0.7612\n",
      "Epoch 56/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0741 - accuracy: 0.9989 - val_loss: 1.0348 - val_accuracy: 0.7550\n",
      "Epoch 57/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0741 - accuracy: 0.9983 - val_loss: 1.0272 - val_accuracy: 0.7587\n",
      "Epoch 58/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0730 - accuracy: 0.9988 - val_loss: 1.0430 - val_accuracy: 0.7425\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Epoch 59/100\n",
      "19/19 [==============================] - 87s 5s/step - loss: 0.0721 - accuracy: 0.9989 - val_loss: 1.0563 - val_accuracy: 0.7512\n",
      "Epoch 60/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0715 - accuracy: 0.9990 - val_loss: 1.0076 - val_accuracy: 0.7612\n",
      "Epoch 61/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0710 - accuracy: 0.9982 - val_loss: 1.0108 - val_accuracy: 0.7650\n",
      "Epoch 62/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0705 - accuracy: 0.9985 - val_loss: 1.0007 - val_accuracy: 0.7587\n",
      "Epoch 63/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0710 - accuracy: 0.9979 - val_loss: 1.0384 - val_accuracy: 0.7550\n",
      "Epoch 64/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0701 - accuracy: 0.9989 - val_loss: 0.9823 - val_accuracy: 0.7625\n",
      "Epoch 65/100\n",
      "19/19 [==============================] - 89s 5s/step - loss: 0.0685 - accuracy: 0.9984 - val_loss: 1.0195 - val_accuracy: 0.7537\n",
      "Epoch 66/100\n",
      "19/19 [==============================] - 91s 5s/step - loss: 0.0674 - accuracy: 0.9985 - val_loss: 1.0250 - val_accuracy: 0.7537\n",
      "Epoch 67/100\n",
      "19/19 [==============================] - 101s 5s/step - loss: 0.0665 - accuracy: 0.9989 - val_loss: 1.0186 - val_accuracy: 0.7625\n",
      "Epoch 68/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0657 - accuracy: 0.9990 - val_loss: 1.0105 - val_accuracy: 0.7537\n",
      "Epoch 69/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0645 - accuracy: 0.9991 - val_loss: 1.0379 - val_accuracy: 0.7512\n",
      "Epoch 70/100\n",
      "19/19 [==============================] - 91s 5s/step - loss: 0.0650 - accuracy: 0.9989 - val_loss: 1.0630 - val_accuracy: 0.7437\n",
      "Epoch 71/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0655 - accuracy: 0.9988 - val_loss: 1.0462 - val_accuracy: 0.7562\n",
      "Epoch 72/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0644 - accuracy: 0.9993 - val_loss: 1.0304 - val_accuracy: 0.7575\n",
      "Epoch 73/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0650 - accuracy: 0.9988 - val_loss: 1.0705 - val_accuracy: 0.7475\n",
      "Epoch 74/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.0642 - accuracy: 0.9990 - val_loss: 1.0450 - val_accuracy: 0.7575\n",
      "Epoch 75/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0640 - accuracy: 0.9989 - val_loss: 1.0307 - val_accuracy: 0.7587\n",
      "Epoch 76/100\n",
      "19/19 [==============================] - 94s 5s/step - loss: 0.0630 - accuracy: 0.9991 - val_loss: 1.0832 - val_accuracy: 0.7500\n",
      "Epoch 77/100\n",
      "19/19 [==============================] - 104s 6s/step - loss: 0.0643 - accuracy: 0.9986 - val_loss: 1.0637 - val_accuracy: 0.7450\n",
      "Epoch 78/100\n",
      "19/19 [==============================] - 105s 6s/step - loss: 0.0635 - accuracy: 0.9993 - val_loss: 1.0380 - val_accuracy: 0.7537\n",
      "Epoch 79/100\n",
      "19/19 [==============================] - 107s 6s/step - loss: 0.0628 - accuracy: 0.9988 - val_loss: 1.0976 - val_accuracy: 0.7462\n",
      "Epoch 80/100\n",
      "19/19 [==============================] - 106s 6s/step - loss: 0.0619 - accuracy: 0.9991 - val_loss: 1.1125 - val_accuracy: 0.7487\n",
      "Epoch 81/100\n",
      "19/19 [==============================] - 93s 5s/step - loss: 0.0619 - accuracy: 0.9990 - val_loss: 1.0940 - val_accuracy: 0.7525\n",
      "Epoch 82/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0634 - accuracy: 0.9986 - val_loss: 1.0990 - val_accuracy: 0.7375\n",
      "Epoch 83/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0643 - accuracy: 0.9984 - val_loss: 1.0887 - val_accuracy: 0.7462\n",
      "Epoch 84/100\n",
      "19/19 [==============================] - 88s 5s/step - loss: 0.0637 - accuracy: 0.9986 - val_loss: 1.0873 - val_accuracy: 0.7462\n",
      "Epoch 85/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0636 - accuracy: 0.9985 - val_loss: 1.0595 - val_accuracy: 0.7462\n",
      "Epoch 86/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0638 - accuracy: 0.9984 - val_loss: 1.0679 - val_accuracy: 0.7450\n",
      "Epoch 87/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0610 - accuracy: 0.9988 - val_loss: 1.0472 - val_accuracy: 0.7462\n",
      "Epoch 88/100\n",
      "19/19 [==============================] - 90s 5s/step - loss: 0.0608 - accuracy: 0.9987 - val_loss: 1.0472 - val_accuracy: 0.7625\n",
      "Epoch 89/100\n",
      "19/19 [==============================] - 89s 5s/step - loss: 0.0606 - accuracy: 0.9988 - val_loss: 1.1315 - val_accuracy: 0.7587\n",
      "Epoch 90/100\n",
      "19/19 [==============================] - 91s 5s/step - loss: 0.0624 - accuracy: 0.9982 - val_loss: 1.0853 - val_accuracy: 0.7600\n",
      "Epoch 91/100\n",
      "19/19 [==============================] - 91s 5s/step - loss: 0.0609 - accuracy: 0.9988 - val_loss: 1.0988 - val_accuracy: 0.7500\n",
      "Epoch 92/100\n",
      "19/19 [==============================] - 99s 5s/step - loss: 0.0629 - accuracy: 0.9980 - val_loss: 1.0956 - val_accuracy: 0.7487\n",
      "Epoch 93/100\n",
      "19/19 [==============================] - 103s 5s/step - loss: 0.0628 - accuracy: 0.9979 - val_loss: 1.1307 - val_accuracy: 0.7437\n",
      "Epoch 94/100\n",
      "19/19 [==============================] - 102s 5s/step - loss: 0.0618 - accuracy: 0.9986 - val_loss: 1.1016 - val_accuracy: 0.7462\n",
      "Epoch 95/100\n",
      "19/19 [==============================] - 101s 5s/step - loss: 0.0614 - accuracy: 0.9980 - val_loss: 1.0941 - val_accuracy: 0.7500\n",
      "Epoch 96/100\n",
      "19/19 [==============================] - 103s 5s/step - loss: 0.0602 - accuracy: 0.9986 - val_loss: 1.1185 - val_accuracy: 0.7537\n",
      "Epoch 97/100\n",
      "19/19 [==============================] - 105s 6s/step - loss: 0.0595 - accuracy: 0.9991 - val_loss: 1.0701 - val_accuracy: 0.7537\n",
      "Epoch 98/100\n",
      "19/19 [==============================] - 107s 6s/step - loss: 0.0579 - accuracy: 0.9991 - val_loss: 1.0821 - val_accuracy: 0.7562\n",
      "Epoch 99/100\n",
      "19/19 [==============================] - 95s 5s/step - loss: 0.0583 - accuracy: 0.9988 - val_loss: 1.1283 - val_accuracy: 0.7575\n",
      "Epoch 100/100\n",
      "19/19 [==============================] - 92s 5s/step - loss: 0.0598 - accuracy: 0.9985 - val_loss: 1.1190 - val_accuracy: 0.7625\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.training.tracking.util.CheckpointLoadStatus at 0x10f8a1490>"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "history = model1.fit([fine_tune_data_4P_train, fine_tune_data_train, fine_tune_mask_train], fine_tune_labels_train, epochs=EPOCHS, batch_size=800, validation_data=([fine_tune_data_4P_test, fine_tune_data_test, fine_tune_mask_test], fine_tune_labels_test), validation_batch_size=800, callbacks=[model_checkpoint_callback])\n",
    "# history = model1.fit(fine_tune_data, fine_tune_labels, epochs=EPOCHS, batch_size=160z0, callbacks=[model_checkpoint_callback])\n",
    "model1.save_weights(\"./Models/Inception-ResNet-kP-ViT/inception-resnet-kp-model-weights\")\n",
    "model1.load_weights('./Models/Inception-ResNet-kP-ViT/inception-resnet-kp-model-weights')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "8e04992f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<tensorflow.python.training.tracking.util.CheckpointLoadStatus at 0x1082bfbb0>"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model1.load_weights('./Models/Inception-ResNet-kP-ViT/auto-Inception-ResNet-FT-model-weight')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "9116a6a6",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_train_loss(history, model_name):  \n",
    "    plt.figure(figsize=(8,6))\n",
    "    plt.plot(history.history['loss'])\n",
    "    plt.plot(history.history['accuracy'])\n",
    "    plt.plot(history.history['val_loss'])\n",
    "    plt.plot(history.history['val_accuracy'])\n",
    "    plt.title(model_name+' loss')\n",
    "    plt.ylabel('loss')\n",
    "    plt.xlabel('epochs')\n",
    "    plt.legend(['train loss', 'train accuracy', 'val loss', 'val accuracy'], loc='upper left')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "c55b5870",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_train_loss(history, 'Fine-Tuned Inception-ResNet-kPiece-ViT Model')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03390fa6",
   "metadata": {},
   "source": [
    "### Model Evaluation"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48701107",
   "metadata": {},
   "source": [
    "#### On Train Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b0594069",
   "metadata": {},
   "source": [
    "Now, after the model is being trained, we will look at the performance of the model on the training set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "cf7ded38",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 14:26:31.344697: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "475/475 [==============================] - 49s 101ms/step\n"
     ]
    }
   ],
   "source": [
    "train_preds = model1.predict([fine_tune_data_4P_train, fine_tune_data_train, fine_tune_mask_train])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "f2a28269",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_pred_labels = np.argmax(train_preds, axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "768775b2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1, 13, 13, ...,  3,  9,  9])"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_pred_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "eff52a3d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "15197"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(train_pred_labels == train_labels['label'])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3fd403fa",
   "metadata": {},
   "source": [
    "Looking at the misclassified images to get a better idea about what features the model is unable to capture."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "8f594a33",
   "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>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9198</th>\n",
       "      <td>9682</td>\n",
       "      <td>./Data/train/train/9682.tif</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12255</th>\n",
       "      <td>12896</td>\n",
       "      <td>./Data/train/train/12896.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14569</th>\n",
       "      <td>15333</td>\n",
       "      <td>./Data/train/train/15333.tif</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          id                        images  label\n",
       "9198    9682   ./Data/train/train/9682.tif     12\n",
       "12255  12896  ./Data/train/train/12896.tif      9\n",
       "14569  15333  ./Data/train/train/15333.tif      7"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mis_train_data = train_labels[train_pred_labels != train_labels['label']]\n",
    "mis_train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "2ea8f949",
   "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>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>fscore</th>\n",
       "      <th>support</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.997959</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998979</td>\n",
       "      <td>978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>925</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.998926</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999463</td>\n",
       "      <td>930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>962</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>998</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998905</td>\n",
       "      <td>0.999452</td>\n",
       "      <td>913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998951</td>\n",
       "      <td>0.999475</td>\n",
       "      <td>953</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>916</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.998976</td>\n",
       "      <td>0.999488</td>\n",
       "      <td>977</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>943</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>945</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    precision    recall    fscore  support\n",
       "0    0.997959  1.000000  0.998979      978\n",
       "1    1.000000  1.000000  1.000000      925\n",
       "2    1.000000  1.000000  1.000000      925\n",
       "3    1.000000  1.000000  1.000000      947\n",
       "4    0.998926  1.000000  0.999463      930\n",
       "5    1.000000  1.000000  1.000000      962\n",
       "6    1.000000  1.000000  1.000000      998\n",
       "7    1.000000  0.998905  0.999452      913\n",
       "8    1.000000  1.000000  1.000000     1005\n",
       "9    1.000000  0.998951  0.999475      953\n",
       "10   1.000000  1.000000  1.000000      916\n",
       "11   1.000000  1.000000  1.000000      955\n",
       "12   1.000000  0.998976  0.999488      977\n",
       "13   1.000000  1.000000  1.000000      943\n",
       "14   1.000000  1.000000  1.000000      928\n",
       "15   1.000000  1.000000  1.000000      945"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision, recall, fscore, support = score(train_labels['label'], train_pred_labels)\n",
    "pd.DataFrame({\n",
    "    'precision': precision,\n",
    "    'recall': recall,\n",
    "    'fscore': fscore,\n",
    "    'support': support\n",
    "})"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b6e51475",
   "metadata": {},
   "source": [
    "#### On Train-Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "14a7854d",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "25/25 [==============================] - 3s 104ms/step\n"
     ]
    }
   ],
   "source": [
    "test_preds = model1.predict([fine_tune_data_4P_test, fine_tune_data_test, fine_tune_mask_test])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "d585fa70",
   "metadata": {},
   "outputs": [],
   "source": [
    "test_pred_labels = np.argmax(test_preds, axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "9a77dc3c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([11,  8,  5,  7,  7,  9, 13,  4, 14,  1,  6, 15,  0, 15, 13, 11,  0,\n",
       "        5,  9, 10,  8, 14,  3, 12,  6,  0,  3,  8,  3,  3, 11,  1,  9,  2,\n",
       "       13,  8,  4,  7, 11,  9,  6, 12, 14,  2,  5, 14,  7, 15, 13,  4, 15,\n",
       "       10,  5, 15, 11, 14, 15,  9,  9,  2, 15,  6, 11, 11,  6,  8,  8,  0,\n",
       "        6,  3,  4,  4, 12, 12,  8,  6,  7,  0,  0,  2, 11, 15, 12, 12,  9,\n",
       "        8, 13, 13, 13, 13, 10, 11,  4, 10,  1,  7, 11,  8, 13, 10,  3, 10,\n",
       "        2,  9, 10, 14,  8,  2,  5, 12,  7,  5,  9,  4, 12,  9, 12, 13, 11,\n",
       "        6,  7,  1,  0, 12,  1,  5, 10,  8, 15, 14,  3, 11,  6,  1,  6,  9,\n",
       "       14, 10,  7, 14, 10,  8, 11, 12,  8, 10,  8,  9,  2, 13,  0,  5,  4,\n",
       "       14,  4,  9,  4, 14,  7, 11,  8,  4,  4, 10,  5,  8, 15,  8,  8,  8,\n",
       "       13,  6,  3, 11,  7, 12, 10, 14, 14,  8,  6, 11,  1, 11,  6, 13,  4,\n",
       "       14, 11,  7, 15,  0,  3,  0, 10, 13, 12, 10, 11, 11, 11,  1,  9, 13,\n",
       "        9,  1, 13, 14, 12,  1,  4,  8,  7,  6, 13,  5,  4,  5,  8,  0,  9,\n",
       "       13, 15,  0,  7,  5,  2,  2,  6,  0,  7,  0,  3, 12,  3,  1,  2, 12,\n",
       "        4,  7, 12, 13, 12,  8, 13, 10,  0,  0, 11,  0, 10, 13, 15, 12,  8,\n",
       "        1, 10, 13,  5,  1,  9, 14,  6, 13, 12, 15, 11,  0, 11, 11, 14,  3,\n",
       "        0, 11,  4,  3,  1,  4,  6,  2,  7,  8, 10, 10,  8, 15,  7,  8, 13,\n",
       "        6,  8, 15,  0,  2,  3,  4, 10,  0,  8,  3,  0,  7, 15, 10,  7,  5,\n",
       "        3, 13, 11, 10,  9,  2, 13,  3, 15,  6,  0,  4, 14, 11, 13,  2, 15,\n",
       "        9, 15, 15, 10,  0, 15,  3, 11, 13, 15, 15,  7, 10,  4, 13, 15,  4,\n",
       "       13,  0,  3,  3,  2,  8,  2,  1, 11,  3,  3,  2, 12, 14,  4,  0,  3,\n",
       "        8,  5,  1,  3,  0,  5,  1, 12,  7, 15, 15,  3,  2, 12, 15,  3,  6,\n",
       "        0, 14, 14,  0,  5,  8,  2,  6, 12, 13,  6,  9,  8,  7, 10,  1,  2,\n",
       "       15, 14,  2, 14, 15,  1,  0,  1,  2, 10,  1,  0, 14,  6,  9,  0,  2,\n",
       "        5,  3, 14,  8, 13, 11,  9, 14, 13,  1,  4,  7, 13, 14,  5, 10,  8,\n",
       "        0,  5, 11,  0,  1,  3,  7,  2,  5,  6, 15,  9, 11, 13,  3, 10,  8,\n",
       "       12,  0,  3,  5,  5,  9,  6, 14,  7,  8,  3,  2, 13, 12,  0, 13, 11,\n",
       "        9,  8, 10, 14, 12, 13, 15, 10,  9,  9, 15, 12,  7, 14,  9, 15,  7,\n",
       "        5, 10,  4,  8,  9,  3,  7,  8,  6, 10,  1, 10,  7, 10, 13, 13, 10,\n",
       "        2,  3,  1, 10,  0,  7,  8, 12,  9,  1,  6,  5, 15, 15,  8,  7, 10,\n",
       "       12,  1,  0, 12, 11,  1,  1,  7,  9, 10, 14,  3,  1, 14,  6, 14,  7,\n",
       "        1, 14,  6, 14,  9,  5, 13, 14,  1,  1,  2,  3,  7,  2,  0,  8,  8,\n",
       "        0,  8,  6, 15, 14,  3,  6, 11,  8, 14,  6,  9,  5,  9, 11, 15,  0,\n",
       "        3, 10,  5, 10, 15,  2, 13,  0, 14,  9,  8,  0,  6, 13,  7, 10, 12,\n",
       "        9, 11,  6, 10,  9,  8, 10, 14, 14,  9,  0, 15,  8,  5,  8,  6,  5,\n",
       "       12,  5,  0, 15, 14,  5,  3,  2, 12,  1, 15, 11, 10,  3, 13,  1,  3,\n",
       "       12, 13,  0,  4,  4,  9,  5,  5, 14, 13,  5,  8,  5,  9, 13,  4, 15,\n",
       "       15,  0,  1, 14, 13,  7,  6, 12,  8,  7,  5, 15,  5,  9,  4,  1, 14,\n",
       "        1, 10,  6,  6,  3,  7,  7,  5,  5, 12,  6,  8,  0,  4,  7,  7,  0,\n",
       "       12,  6, 12,  0,  3, 11,  2,  3,  5,  4, 13,  1, 12,  5, 14, 15,  4,\n",
       "       12,  5, 10,  7,  2,  2, 12,  8, 11,  6,  9,  2, 10, 13,  7,  6,  3,\n",
       "       13,  3,  2, 14,  0,  8, 12, 12,  8, 13, 13, 14,  4, 10,  2,  5, 14,\n",
       "       12, 13, 13,  4, 12,  0, 11,  0,  8, 10, 13,  8,  4,  9,  2,  8,  0,\n",
       "        1,  2, 14, 15,  0,  8,  9, 11,  0,  8,  9,  7,  4, 11, 13, 13,  3,\n",
       "        7,  7,  3,  8, 12,  2, 12, 15,  7, 10, 13, 14, 12,  6, 14,  0, 10,\n",
       "       11,  3, 12, 13,  5,  7,  2,  8,  5, 13,  2,  9,  8, 14, 12, 14, 12,\n",
       "        0,  5,  8, 15,  4, 11,  0,  0,  9, 12,  5,  1, 10,  2,  2, 11, 10,\n",
       "        6])"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_pred_labels"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "8bce361f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "605"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.sum(test_pred_labels == test_labels['label'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "e07d40ba",
   "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>id</th>\n",
       "      <th>images</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>33</td>\n",
       "      <td>./Data/train/train/33.tif</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>41</td>\n",
       "      <td>./Data/train/train/41.tif</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>65</td>\n",
       "      <td>./Data/train/train/65.tif</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>100</td>\n",
       "      <td>./Data/train/train/100.tif</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>179</td>\n",
       "      <td>./Data/train/train/179.tif</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>782</th>\n",
       "      <td>15582</td>\n",
       "      <td>./Data/train/train/15582.tif</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>788</th>\n",
       "      <td>15687</td>\n",
       "      <td>./Data/train/train/15687.tif</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>791</th>\n",
       "      <td>15719</td>\n",
       "      <td>./Data/train/train/15719.tif</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>793</th>\n",
       "      <td>15762</td>\n",
       "      <td>./Data/train/train/15762.tif</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>798</th>\n",
       "      <td>15863</td>\n",
       "      <td>./Data/train/train/15863.tif</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>195 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id                        images  label\n",
       "2       33     ./Data/train/train/33.tif      6\n",
       "3       41     ./Data/train/train/41.tif      1\n",
       "4       65     ./Data/train/train/65.tif     14\n",
       "7      100    ./Data/train/train/100.tif     10\n",
       "12     179    ./Data/train/train/179.tif      7\n",
       "..     ...                           ...    ...\n",
       "782  15582  ./Data/train/train/15582.tif      3\n",
       "788  15687  ./Data/train/train/15687.tif     12\n",
       "791  15719  ./Data/train/train/15719.tif      9\n",
       "793  15762  ./Data/train/train/15762.tif     13\n",
       "798  15863  ./Data/train/train/15863.tif     15\n",
       "\n",
       "[195 rows x 3 columns]"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mis_train_data = test_labels[test_pred_labels != test_labels['label']]\n",
    "mis_train_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "0f0b85fa",
   "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>precision</th>\n",
       "      <th>recall</th>\n",
       "      <th>fscore</th>\n",
       "      <th>support</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.740741</td>\n",
       "      <td>0.701754</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.585366</td>\n",
       "      <td>0.545455</td>\n",
       "      <td>0.564706</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.883721</td>\n",
       "      <td>0.883721</td>\n",
       "      <td>0.883721</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0.895833</td>\n",
       "      <td>0.934783</td>\n",
       "      <td>0.914894</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.891892</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>0.904110</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0.583333</td>\n",
       "      <td>0.560000</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>0.886364</td>\n",
       "      <td>0.847826</td>\n",
       "      <td>0.866667</td>\n",
       "      <td>46</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>0.795918</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.772277</td>\n",
       "      <td>52</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>0.815385</td>\n",
       "      <td>0.981481</td>\n",
       "      <td>0.890756</td>\n",
       "      <td>54</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>0.787234</td>\n",
       "      <td>0.787234</td>\n",
       "      <td>0.787234</td>\n",
       "      <td>47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>0.666667</td>\n",
       "      <td>0.620690</td>\n",
       "      <td>0.642857</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>0.787234</td>\n",
       "      <td>0.660714</td>\n",
       "      <td>0.718447</td>\n",
       "      <td>56</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>0.622642</td>\n",
       "      <td>0.767442</td>\n",
       "      <td>0.687500</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>0.754098</td>\n",
       "      <td>0.718750</td>\n",
       "      <td>0.736000</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>0.907407</td>\n",
       "      <td>0.765625</td>\n",
       "      <td>0.830508</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>0.612245</td>\n",
       "      <td>0.697674</td>\n",
       "      <td>0.652174</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    precision    recall    fscore  support\n",
       "0    0.666667  0.740741  0.701754       54\n",
       "1    0.585366  0.545455  0.564706       44\n",
       "2    0.883721  0.883721  0.883721       43\n",
       "3    0.895833  0.934783  0.914894       46\n",
       "4    0.891892  0.916667  0.904110       36\n",
       "5    0.583333  0.560000  0.571429       50\n",
       "6    0.886364  0.847826  0.866667       46\n",
       "7    0.795918  0.750000  0.772277       52\n",
       "8    0.815385  0.981481  0.890756       54\n",
       "9    0.787234  0.787234  0.787234       47\n",
       "10   0.666667  0.620690  0.642857       58\n",
       "11   0.787234  0.660714  0.718447       56\n",
       "12   0.622642  0.767442  0.687500       43\n",
       "13   0.754098  0.718750  0.736000       64\n",
       "14   0.907407  0.765625  0.830508       64\n",
       "15   0.612245  0.697674  0.652174       43"
      ]
     },
     "execution_count": 73,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "precision, recall, fscore, support = score(test_labels['label'], test_pred_labels)\n",
    "pd.DataFrame({\n",
    "    'precision': precision,\n",
    "    'recall': recall,\n",
    "    'fscore': fscore,\n",
    "    'support': support\n",
    "})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "ffd70520",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7578145842452271"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "np.mean(fscore)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "067c5096",
   "metadata": {},
   "source": [
    "#### On Validation Data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e665591",
   "metadata": {},
   "source": [
    "Generating Predictions for the Validation Set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "20d57a31",
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "2022-10-09 14:27:46.704591: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:113] Plugin optimizer for device_type GPU is enabled.\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "29/29 [==============================] - 4s 117ms/step\n"
     ]
    }
   ],
   "source": [
    "valid_preds = model1.predict([fine_tune_data_4P_validation, fine_tune_data_validation, mask_validation])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "db7d1256",
   "metadata": {},
   "outputs": [],
   "source": [
    "valid_pred_labels = np.argmax(valid_preds, axis=-1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "315c93f9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 4,  6,  4,  8,  3,  8, 10, 11,  2,  4,  0, 15,  5, 12, 13,  4,  1,\n",
       "        1,  0, 12, 14,  9, 13,  6,  1,  2,  7,  3,  8,  1,  0, 14,  8, 10,\n",
       "       14,  6, 11,  8,  6, 10, 12,  7, 12,  3,  2, 10,  3,  2,  6,  2,  3,\n",
       "        0, 10,  0, 10,  9,  2,  8,  1, 11,  2,  7,  9,  5, 12,  8, 11,  7,\n",
       "        8,  4,  5, 12,  4,  8,  4,  6,  4,  7,  5,  0,  9,  0, 13,  8,  7,\n",
       "       12, 11, 10, 15, 10, 12, 12,  1, 12, 11,  4, 10, 10,  0,  6,  6, 11,\n",
       "        1, 12,  7,  1, 10,  6, 10,  1,  7,  0, 11, 14,  2,  4,  1, 14,  7,\n",
       "       11, 12, 12,  3,  9, 12, 10,  0,  6,  4,  6,  8,  1,  0,  9, 10,  3,\n",
       "        8, 11,  5,  5,  6,  7,  2,  3, 13,  9,  9, 12,  0, 14, 10, 13, 13,\n",
       "        3, 12, 10,  0,  7,  5, 14,  3, 11,  2, 13,  7, 14,  4,  4,  7,  9,\n",
       "        6, 11,  4,  7,  4,  0,  1,  0,  0,  8,  9, 11,  3,  2,  3,  3, 13,\n",
       "        8, 11,  2,  9,  0,  9,  2,  4, 14,  2,  9,  8,  7, 11, 15,  1,  0,\n",
       "       15, 13,  4,  1,  4,  4, 10,  6,  9, 10,  4,  4, 13,  3, 10, 13, 13,\n",
       "        6,  0,  8, 12,  0,  5,  4,  3, 13,  5, 13,  8, 11,  2,  6,  7,  5,\n",
       "       13, 12,  9,  3, 14, 13,  7, 10,  8,  0,  4,  4,  9,  8,  9, 11,  7,\n",
       "        8,  9, 15,  6,  0, 14,  5,  0,  6,  9,  7, 12,  5, 15,  4,  3,  2,\n",
       "        1,  5,  8,  5,  9,  2, 10, 15,  8, 10,  3,  3, 13, 10,  2, 11, 10,\n",
       "        9, 11,  1,  6,  2, 14, 11, 15, 11,  4, 12,  3,  2, 15, 10, 14,  8,\n",
       "       12,  2, 10,  6, 15,  0, 10,  0, 10,  7,  6,  8, 13,  4, 13,  6,  6,\n",
       "        4, 13,  5,  5,  0, 13, 12,  5, 10, 11, 13,  7,  2, 14, 12, 10,  4,\n",
       "       15, 13,  4,  9, 10,  4, 11,  5, 13,  6,  3, 12,  5, 14,  4, 12, 13,\n",
       "       15, 10,  8,  5, 15,  3,  0,  3, 10,  7, 10,  3,  7, 12,  9, 10,  2,\n",
       "        1, 11,  9, 10,  7,  7,  8, 13,  6,  3,  4, 10,  2, 11,  0, 11,  0,\n",
       "       10, 14,  4, 14, 15, 10,  1,  7,  1, 10,  2, 12,  5, 12, 10,  0,  1,\n",
       "        3, 14, 10,  6, 13,  1,  2,  1, 14,  4,  3,  3,  2,  0, 15, 12,  4,\n",
       "       11, 13,  9,  2,  6,  3,  2, 15,  9,  0,  2,  8,  6, 10, 12,  6,  1,\n",
       "       14, 12,  0,  7,  3,  7, 14,  5, 10, 11,  6,  7, 11, 15,  4, 11,  2,\n",
       "        5,  4,  3, 10,  0,  8, 10,  4,  4,  1,  4,  9,  7, 12,  6,  4,  4,\n",
       "        9,  1, 15, 11,  2, 13,  8,  4,  2, 12, 12, 14,  6, 15, 12,  9,  9,\n",
       "        5,  0, 13,  2,  3, 11, 10,  7,  4, 13,  8,  2, 12,  8, 10, 15,  7,\n",
       "       14, 12,  8, 11,  8,  0,  2, 13, 12,  3,  8,  4,  5,  4,  6,  0, 10,\n",
       "        5, 15,  3, 12, 10,  9, 13,  3,  5, 14, 12, 13, 10,  9,  8,  5, 11,\n",
       "        4,  8,  8,  7,  5,  9, 13,  8,  8,  3,  4,  4,  1,  2,  3,  0, 13,\n",
       "        7,  0, 11, 15, 11,  1, 15,  6, 15,  4,  4,  7,  1,  3,  5, 15,  4,\n",
       "        5, 12,  8, 13, 11,  4,  8,  2, 10,  9,  7,  2,  0,  2, 13,  9,  4,\n",
       "        6,  6, 12,  0,  6,  8,  1,  5, 13,  2,  2,  4, 12,  8, 10,  3, 14,\n",
       "        4, 13, 11, 14,  6,  3, 13,  7,  8, 12,  4, 14, 15, 14, 15, 12,  3,\n",
       "       14,  6,  1,  7, 14,  7, 11, 12, 10,  5,  2, 12,  3,  4,  5, 14,  9,\n",
       "        9,  4, 12,  2, 10,  2, 15,  0, 11,  6,  3,  8,  9,  5,  3,  7,  3,\n",
       "        2, 13, 13, 12,  5, 11,  2, 10,  4,  0, 12, 11, 11,  2, 13,  6, 14,\n",
       "       13,  9,  3,  2, 10,  0,  8,  8,  4, 15,  6, 13, 10,  3, 13, 13, 10,\n",
       "       12,  1,  4, 10,  8,  0, 15,  4,  5, 11,  3, 12,  6,  7,  9,  9,  8,\n",
       "        3,  3, 11, 10,  1,  0,  6, 13, 14,  9, 14,  6, 11, 11,  0, 10, 11,\n",
       "        5,  7,  4,  9,  4,  3, 12, 14,  5,  8, 10, 15, 13, 15, 15,  2,  0,\n",
       "        0,  6,  0,  9,  1,  7,  6,  1,  6,  8,  5,  2,  5,  4,  0,  0,  5,\n",
       "        6,  0, 14,  8,  4, 13,  1, 10, 11,  0, 11,  2,  7, 12,  0,  1,  7,\n",
       "        2,  3,  4,  8, 14, 10, 14,  2, 12,  2,  1,  0, 14, 11,  2, 11,  2,\n",
       "       10,  6,  4, 15, 10,  7,  2,  2,  8,  9,  4, 10,  7, 12,  5,  3,  2,\n",
       "       13,  4,  9,  5,  9, 12,  5, 14,  5,  9,  2,  4, 15,  0,  5, 11, 10,\n",
       "        6,  3,  8,  0,  0, 15,  4,  8,  9,  3,  4,  8, 14, 13, 12, 13, 11,\n",
       "        2, 12,  5, 12,  4,  5,  6,  0, 10,  5, 10,  0,  5,  6, 14,  5,  0,\n",
       "        8, 11,  8,  8, 15,  9,  0, 11,  3,  3,  3,  0,  2, 10, 11,  6, 15,\n",
       "       12, 12, 10,  2, 11,  4,  0,  7,  6, 14,  8,  7, 12,  8,  3,  4])"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "valid_pred_labels"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "2bece594",
   "metadata": {},
   "source": [
    "### Submission"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "304c47de",
   "metadata": {},
   "source": [
    "For Kaggle Submission"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "e7ca9d19",
   "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>id</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>17801</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>17802</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17803</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>17804</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>17805</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>895</th>\n",
       "      <td>18696</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>896</th>\n",
       "      <td>18697</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>897</th>\n",
       "      <td>18698</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>898</th>\n",
       "      <td>18699</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>899</th>\n",
       "      <td>18700</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>900 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        id  label\n",
       "0    17801      4\n",
       "1    17802      6\n",
       "2    17803      4\n",
       "3    17804      8\n",
       "4    17805      3\n",
       "..     ...    ...\n",
       "895  18696      7\n",
       "896  18697     12\n",
       "897  18698      8\n",
       "898  18699      3\n",
       "899  18700      4\n",
       "\n",
       "[900 rows x 2 columns]"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submission_data = pd.DataFrame({'id':list(range(17801, 18701))})\n",
    "submission_data['label'] = valid_pred_labels\n",
    "submission_data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "6eea7d9c",
   "metadata": {},
   "outputs": [],
   "source": [
    "submission_data.to_csv('./Submissions/submission-inception-resnet-kPiece-ViT.csv', index=False)"
   ]
  }
 ],
 "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": "3.8.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}