model_statistics.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style\n",
    "import seaborn as sns\n",
    "import random\n",
    "\n",
    "%matplotlib inline\n",
    "style.use('fivethirtyeight')\n",
    "sns.set(style='whitegrid', color_codes=True)\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, roc_curve, roc_auc_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras import backend as K\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten, Activation\n",
    "from keras.layers import Conv2D, MaxPooling2D, BatchNormalization, GlobalAveragePooling2D\n",
    "from tensorflow.keras.optimizers import Adam, SGD, Adagrad, Adadelta, RMSprop\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.callbacks import ModelCheckpoint, EarlyStopping, TensorBoard, CSVLogger, ReduceLROnPlateau\n",
    "\n",
    "import tensorflow as tf\n",
    "import random as rn\n",
    "\n",
    "import os\n",
    "from random import shuffle\n",
    "from zipfile import ZipFile\n",
    "from keras.models import load_model\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "ename": "TypeError",
     "evalue": "Descriptors cannot not be created directly.\nIf this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.\nIf you cannot immediately regenerate your protos, some other possible workarounds are:\n 1. Downgrade the protobuf package to 3.20.x or lower.\n 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).\n\nMore information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[2], line 19\u001b[0m\n\u001b[0;32m     16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msklearn\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mmodel_selection\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m GridSearchCV\n\u001b[0;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01msklearn\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpreprocessing\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m LabelEncoder\n\u001b[1;32m---> 19\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpreprocessing\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mimage\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m ImageDataGenerator\n\u001b[0;32m     20\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m backend \u001b[38;5;28;01mas\u001b[39;00m K\n\u001b[0;32m     21\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mmodels\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Sequential\n",
      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\lib\\site-packages\\keras\\__init__.py:20\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# Copyright 2015 The TensorFlow Authors. All Rights Reserved.\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m#\u001b[39;00m\n\u001b[0;32m      3\u001b[0m \u001b[38;5;66;03m# Licensed under the Apache License, Version 2.0 (the \"License\");\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     13\u001b[0m \u001b[38;5;66;03m# limitations under the License.\u001b[39;00m\n\u001b[0;32m     14\u001b[0m \u001b[38;5;66;03m# ==============================================================================\u001b[39;00m\n\u001b[0;32m     15\u001b[0m \u001b[38;5;124;03m\"\"\"Implementation of the Keras API, the high-level API of TensorFlow.\u001b[39;00m\n\u001b[0;32m     16\u001b[0m \n\u001b[0;32m     17\u001b[0m \u001b[38;5;124;03mDetailed documentation and user guides are available at\u001b[39;00m\n\u001b[0;32m     18\u001b[0m \u001b[38;5;124;03m[keras.io](https://keras.io).\u001b[39;00m\n\u001b[0;32m     19\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m---> 20\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m distribute\n\u001b[0;32m     21\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m models\n\u001b[0;32m     22\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mengine\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01minput_layer\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m Input\n",
      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\lib\\site-packages\\keras\\distribute\\__init__.py:18\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# Copyright 2019 The TensorFlow Authors. All Rights Reserved.\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m#\u001b[39;00m\n\u001b[0;32m      3\u001b[0m \u001b[38;5;66;03m# Licensed under the Apache License, Version 2.0 (the \"License\");\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     13\u001b[0m \u001b[38;5;66;03m# limitations under the License.\u001b[39;00m\n\u001b[0;32m     14\u001b[0m \u001b[38;5;66;03m# ==============================================================================\u001b[39;00m\n\u001b[0;32m     15\u001b[0m \u001b[38;5;124;03m\"\"\"Keras' Distribution Strategy library.\"\"\"\u001b[39;00m\n\u001b[1;32m---> 18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mkeras\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mdistribute\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m sidecar_evaluator\n",
      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\lib\\site-packages\\keras\\distribute\\sidecar_evaluator.py:17\u001b[0m\n\u001b[0;32m      1\u001b[0m \u001b[38;5;66;03m# Copyright 2020 The TensorFlow Authors. All Rights Reserved.\u001b[39;00m\n\u001b[0;32m      2\u001b[0m \u001b[38;5;66;03m#\u001b[39;00m\n\u001b[0;32m      3\u001b[0m \u001b[38;5;66;03m# Licensed under the Apache License, Version 2.0 (the \"License\");\u001b[39;00m\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     13\u001b[0m \u001b[38;5;66;03m# limitations under the License.\u001b[39;00m\n\u001b[0;32m     14\u001b[0m \u001b[38;5;66;03m# ==============================================================================\u001b[39;00m\n\u001b[0;32m     15\u001b[0m \u001b[38;5;124;03m\"\"\"Python module for evaluation loop.\"\"\"\u001b[39;00m\n\u001b[1;32m---> 17\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcompat\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mv2\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mtf\u001b[39;00m\n\u001b[0;32m     19\u001b[0m \u001b[38;5;66;03m# isort: off\u001b[39;00m\n\u001b[0;32m     20\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mplatform\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m tf_logging \u001b[38;5;28;01mas\u001b[39;00m logging\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\__init__.py:37\u001b[0m\n\u001b[0;32m     34\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01msys\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01m_sys\u001b[39;00m\n\u001b[0;32m     35\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtyping\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01m_typing\u001b[39;00m\n\u001b[1;32m---> 37\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mtools\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m module_util \u001b[38;5;28;01mas\u001b[39;00m _module_util\n\u001b[0;32m     38\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mutil\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mlazy_loader\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m LazyLoader \u001b[38;5;28;01mas\u001b[39;00m _LazyLoader\n\u001b[0;32m     40\u001b[0m \u001b[38;5;66;03m# Make sure code inside the TensorFlow codebase can use tf2.enabled() at import.\u001b[39;00m\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\python\\__init__.py:37\u001b[0m\n\u001b[0;32m     29\u001b[0m \u001b[38;5;66;03m# We aim to keep this file minimal and ideally remove completely.\u001b[39;00m\n\u001b[0;32m     30\u001b[0m \u001b[38;5;66;03m# If you are adding a new file with @tf_export decorators,\u001b[39;00m\n\u001b[0;32m     31\u001b[0m \u001b[38;5;66;03m# import it in modules_with_exports.py instead.\u001b[39;00m\n\u001b[0;32m     32\u001b[0m \n\u001b[0;32m     33\u001b[0m \u001b[38;5;66;03m# go/tf-wildcard-import\u001b[39;00m\n\u001b[0;32m     34\u001b[0m \u001b[38;5;66;03m# pylint: disable=wildcard-import,g-bad-import-order,g-import-not-at-top\u001b[39;00m\n\u001b[0;32m     36\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m pywrap_tensorflow \u001b[38;5;28;01mas\u001b[39;00m _pywrap_tensorflow\n\u001b[1;32m---> 37\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01meager\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m context\n\u001b[0;32m     39\u001b[0m \u001b[38;5;66;03m# pylint: enable=wildcard-import\u001b[39;00m\n\u001b[0;32m     40\u001b[0m \n\u001b[0;32m     41\u001b[0m \u001b[38;5;66;03m# Bring in subpackages.\u001b[39;00m\n\u001b[0;32m     42\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mpython\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m data\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\python\\eager\\context.py:29\u001b[0m\n\u001b[0;32m     26\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mnumpy\u001b[39;00m \u001b[38;5;28;01mas\u001b[39;00m \u001b[38;5;21;01mnp\u001b[39;00m\n\u001b[0;32m     27\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01msix\u001b[39;00m\n\u001b[1;32m---> 29\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m function_pb2\n\u001b[0;32m     30\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mprotobuf\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m config_pb2\n\u001b[0;32m     31\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mprotobuf\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m coordination_config_pb2\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\core\\framework\\function_pb2.py:16\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[38;5;66;03m# @@protoc_insertion_point(imports)\u001b[39;00m\n\u001b[0;32m     13\u001b[0m _sym_db \u001b[38;5;241m=\u001b[39m _symbol_database\u001b[38;5;241m.\u001b[39mDefault()\n\u001b[1;32m---> 16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m attr_value_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_attr__value__pb2\n\u001b[0;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m node_def_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_node__def__pb2\n\u001b[0;32m     18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m op_def_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_op__def__pb2\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\core\\framework\\attr_value_pb2.py:16\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[38;5;66;03m# @@protoc_insertion_point(imports)\u001b[39;00m\n\u001b[0;32m     13\u001b[0m _sym_db \u001b[38;5;241m=\u001b[39m _symbol_database\u001b[38;5;241m.\u001b[39mDefault()\n\u001b[1;32m---> 16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m tensor_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_tensor__pb2\n\u001b[0;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m tensor_shape_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2\n\u001b[0;32m     18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m types_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_types__pb2\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\core\\framework\\tensor_pb2.py:16\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[38;5;66;03m# @@protoc_insertion_point(imports)\u001b[39;00m\n\u001b[0;32m     13\u001b[0m _sym_db \u001b[38;5;241m=\u001b[39m _symbol_database\u001b[38;5;241m.\u001b[39mDefault()\n\u001b[1;32m---> 16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m resource_handle_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_resource__handle__pb2\n\u001b[0;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m tensor_shape_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2\n\u001b[0;32m     18\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m types_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_types__pb2\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\core\\framework\\resource_handle_pb2.py:16\u001b[0m\n\u001b[0;32m     11\u001b[0m \u001b[38;5;66;03m# @@protoc_insertion_point(imports)\u001b[39;00m\n\u001b[0;32m     13\u001b[0m _sym_db \u001b[38;5;241m=\u001b[39m _symbol_database\u001b[38;5;241m.\u001b[39mDefault()\n\u001b[1;32m---> 16\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m tensor_shape_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2\n\u001b[0;32m     17\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtensorflow\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mcore\u001b[39;00m\u001b[38;5;21;01m.\u001b[39;00m\u001b[38;5;21;01mframework\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m types_pb2 \u001b[38;5;28;01mas\u001b[39;00m tensorflow_dot_core_dot_framework_dot_types__pb2\n\u001b[0;32m     20\u001b[0m DESCRIPTOR \u001b[38;5;241m=\u001b[39m _descriptor\u001b[38;5;241m.\u001b[39mFileDescriptor(\n\u001b[0;32m     21\u001b[0m   name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow/core/framework/resource_handle.proto\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m     22\u001b[0m   package\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     26\u001b[0m   ,\n\u001b[0;32m     27\u001b[0m   dependencies\u001b[38;5;241m=\u001b[39m[tensorflow_dot_core_dot_framework_dot_tensor__shape__pb2\u001b[38;5;241m.\u001b[39mDESCRIPTOR,tensorflow_dot_core_dot_framework_dot_types__pb2\u001b[38;5;241m.\u001b[39mDESCRIPTOR,])\n",
      "File \u001b[1;32m~\\AppData\\Roaming\\Python\\Python310\\site-packages\\tensorflow\\core\\framework\\tensor_shape_pb2.py:36\u001b[0m\n\u001b[0;32m     13\u001b[0m _sym_db \u001b[38;5;241m=\u001b[39m _symbol_database\u001b[38;5;241m.\u001b[39mDefault()\n\u001b[0;32m     18\u001b[0m DESCRIPTOR \u001b[38;5;241m=\u001b[39m _descriptor\u001b[38;5;241m.\u001b[39mFileDescriptor(\n\u001b[0;32m     19\u001b[0m   name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow/core/framework/tensor_shape.proto\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m     20\u001b[0m   package\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m     23\u001b[0m   serialized_pb\u001b[38;5;241m=\u001b[39m_b(\u001b[38;5;124m'\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124m,tensorflow/core/framework/tensor_shape.proto\u001b[39m\u001b[38;5;130;01m\\x12\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;124mtensorflow\u001b[39m\u001b[38;5;130;01m\\\"\u001b[39;00m\u001b[38;5;124mz\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x10\u001b[39;00m\u001b[38;5;124mTensorShapeProto\u001b[39m\u001b[38;5;130;01m\\x12\u001b[39;00m\u001b[38;5;124m-\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x03\u001b[39;00m\u001b[38;5;130;01m\\x64\u001b[39;00m\u001b[38;5;124mim\u001b[39m\u001b[38;5;130;01m\\x18\u001b[39;00m\u001b[38;5;130;01m\\x02\u001b[39;00m\u001b[38;5;124m \u001b[39m\u001b[38;5;130;01m\\x03\u001b[39;00m\u001b[38;5;124m(\u001b[39m\u001b[38;5;130;01m\\x0b\u001b[39;00m\u001b[38;5;130;01m\\x32\u001b[39;00m\u001b[38;5;124m .tensorflow.TensorShapeProto.Dim\u001b[39m\u001b[38;5;130;01m\\x12\u001b[39;00m\u001b[38;5;130;01m\\x14\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x0c\u001b[39;00m\u001b[38;5;124munknown_rank\u001b[39m\u001b[38;5;130;01m\\x18\u001b[39;00m\u001b[38;5;130;01m\\x03\u001b[39;00m\u001b[38;5;124m \u001b[39m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;124m(\u001b[39m\u001b[38;5;130;01m\\x08\u001b[39;00m\u001b[38;5;130;01m\\x1a\u001b[39;00m\u001b[38;5;124m!\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x03\u001b[39;00m\u001b[38;5;130;01m\\x44\u001b[39;00m\u001b[38;5;124mim\u001b[39m\u001b[38;5;130;01m\\x12\u001b[39;00m\u001b[38;5;130;01m\\x0c\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x04\u001b[39;00m\u001b[38;5;124msize\u001b[39m\u001b[38;5;130;01m\\x18\u001b[39;00m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;124m \u001b[39m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;124m(\u001b[39m\u001b[38;5;130;01m\\x03\u001b[39;00m\u001b[38;5;130;01m\\x12\u001b[39;00m\u001b[38;5;130;01m\\x0c\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x04\u001b[39;00m\u001b[38;5;124mname\u001b[39m\u001b[38;5;130;01m\\x18\u001b[39;00m\u001b[38;5;130;01m\\x02\u001b[39;00m\u001b[38;5;124m \u001b[39m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;124m(\u001b[39m\u001b[38;5;130;01m\\t\u001b[39;00m\u001b[38;5;124mB\u001b[39m\u001b[38;5;130;01m\\x87\u001b[39;00m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\x18\u001b[39;00m\u001b[38;5;124morg.tensorflow.frameworkB\u001b[39m\u001b[38;5;130;01m\\x11\u001b[39;00m\u001b[38;5;124mTensorShapeProtosP\u001b[39m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;124mZSgithub.com/tensorflow/tensorflow/tensorflow/go/core/framework/tensor_shape_go_proto\u001b[39m\u001b[38;5;130;01m\\xf8\u001b[39;00m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;130;01m\\x01\u001b[39;00m\u001b[38;5;130;01m\\x62\u001b[39;00m\u001b[38;5;130;01m\\x06\u001b[39;00m\u001b[38;5;124mproto3\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m     24\u001b[0m )\n\u001b[0;32m     29\u001b[0m _TENSORSHAPEPROTO_DIM \u001b[38;5;241m=\u001b[39m _descriptor\u001b[38;5;241m.\u001b[39mDescriptor(\n\u001b[0;32m     30\u001b[0m   name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mDim\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m     31\u001b[0m   full_name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow.TensorShapeProto.Dim\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m     32\u001b[0m   filename\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m     33\u001b[0m   file\u001b[38;5;241m=\u001b[39mDESCRIPTOR,\n\u001b[0;32m     34\u001b[0m   containing_type\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m     35\u001b[0m   fields\u001b[38;5;241m=\u001b[39m[\n\u001b[1;32m---> 36\u001b[0m     \u001b[43m_descriptor\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mFieldDescriptor\u001b[49m\u001b[43m(\u001b[49m\n\u001b[0;32m     37\u001b[0m \u001b[43m      \u001b[49m\u001b[43mname\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43msize\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfull_name\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[38;5;124;43mtensorflow.TensorShapeProto.Dim.size\u001b[39;49m\u001b[38;5;124;43m'\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mindex\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[0;32m     38\u001b[0m \u001b[43m      \u001b[49m\u001b[43mnumber\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[38;5;28;43mtype\u001b[39;49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m3\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcpp_type\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m2\u001b[39;49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mlabel\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m1\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[0;32m     39\u001b[0m \u001b[43m      \u001b[49m\u001b[43mhas_default_value\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdefault_value\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;241;43m0\u001b[39;49m\u001b[43m,\u001b[49m\n\u001b[0;32m     40\u001b[0m \u001b[43m      \u001b[49m\u001b[43mmessage_type\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43menum_type\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mcontaining_type\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[0;32m     41\u001b[0m \u001b[43m      \u001b[49m\u001b[43mis_extension\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mFalse\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mextension_scope\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\n\u001b[0;32m     42\u001b[0m \u001b[43m      \u001b[49m\u001b[43mserialized_options\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[38;5;28;43;01mNone\u001b[39;49;00m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mfile\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mDESCRIPTOR\u001b[49m\u001b[43m)\u001b[49m,\n\u001b[0;32m     43\u001b[0m     _descriptor\u001b[38;5;241m.\u001b[39mFieldDescriptor(\n\u001b[0;32m     44\u001b[0m       name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mname\u001b[39m\u001b[38;5;124m'\u001b[39m, full_name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow.TensorShapeProto.Dim.name\u001b[39m\u001b[38;5;124m'\u001b[39m, index\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m,\n\u001b[0;32m     45\u001b[0m       number\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m2\u001b[39m, \u001b[38;5;28mtype\u001b[39m\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m9\u001b[39m, cpp_type\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m9\u001b[39m, label\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1\u001b[39m,\n\u001b[0;32m     46\u001b[0m       has_default_value\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m, default_value\u001b[38;5;241m=\u001b[39m_b(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m\"\u001b[39m)\u001b[38;5;241m.\u001b[39mdecode(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mutf-8\u001b[39m\u001b[38;5;124m'\u001b[39m),\n\u001b[0;32m     47\u001b[0m       message_type\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, enum_type\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, containing_type\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m     48\u001b[0m       is_extension\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m, extension_scope\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m     49\u001b[0m       serialized_options\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, file\u001b[38;5;241m=\u001b[39mDESCRIPTOR),\n\u001b[0;32m     50\u001b[0m   ],\n\u001b[0;32m     51\u001b[0m   extensions\u001b[38;5;241m=\u001b[39m[\n\u001b[0;32m     52\u001b[0m   ],\n\u001b[0;32m     53\u001b[0m   nested_types\u001b[38;5;241m=\u001b[39m[],\n\u001b[0;32m     54\u001b[0m   enum_types\u001b[38;5;241m=\u001b[39m[\n\u001b[0;32m     55\u001b[0m   ],\n\u001b[0;32m     56\u001b[0m   serialized_options\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m     57\u001b[0m   is_extendable\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mFalse\u001b[39;00m,\n\u001b[0;32m     58\u001b[0m   syntax\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mproto3\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m     59\u001b[0m   extension_ranges\u001b[38;5;241m=\u001b[39m[],\n\u001b[0;32m     60\u001b[0m   oneofs\u001b[38;5;241m=\u001b[39m[\n\u001b[0;32m     61\u001b[0m   ],\n\u001b[0;32m     62\u001b[0m   serialized_start\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m149\u001b[39m,\n\u001b[0;32m     63\u001b[0m   serialized_end\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m182\u001b[39m,\n\u001b[0;32m     64\u001b[0m )\n\u001b[0;32m     66\u001b[0m _TENSORSHAPEPROTO \u001b[38;5;241m=\u001b[39m _descriptor\u001b[38;5;241m.\u001b[39mDescriptor(\n\u001b[0;32m     67\u001b[0m   name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mTensorShapeProto\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[0;32m     68\u001b[0m   full_name\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mtensorflow.TensorShapeProto\u001b[39m\u001b[38;5;124m'\u001b[39m,\n\u001b[1;32m   (...)\u001b[0m\n\u001b[0;32m    100\u001b[0m   serialized_end\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m182\u001b[39m,\n\u001b[0;32m    101\u001b[0m )\n\u001b[0;32m    103\u001b[0m _TENSORSHAPEPROTO_DIM\u001b[38;5;241m.\u001b[39mcontaining_type \u001b[38;5;241m=\u001b[39m _TENSORSHAPEPROTO\n",
      "File \u001b[1;32mc:\\Users\\User\\anaconda3\\lib\\site-packages\\google\\protobuf\\descriptor.py:561\u001b[0m, in \u001b[0;36mFieldDescriptor.__new__\u001b[1;34m(cls, name, full_name, index, number, type, cpp_type, label, default_value, message_type, enum_type, containing_type, is_extension, extension_scope, options, serialized_options, has_default_value, containing_oneof, json_name, file, create_key)\u001b[0m\n\u001b[0;32m    555\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21m__new__\u001b[39m(\u001b[38;5;28mcls\u001b[39m, name, full_name, index, number, \u001b[38;5;28mtype\u001b[39m, cpp_type, label,\n\u001b[0;32m    556\u001b[0m             default_value, message_type, enum_type, containing_type,\n\u001b[0;32m    557\u001b[0m             is_extension, extension_scope, options\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m    558\u001b[0m             serialized_options\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m    559\u001b[0m             has_default_value\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m, containing_oneof\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, json_name\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m,\n\u001b[0;32m    560\u001b[0m             file\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m, create_key\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mNone\u001b[39;00m):  \u001b[38;5;66;03m# pylint: disable=redefined-builtin\u001b[39;00m\n\u001b[1;32m--> 561\u001b[0m   \u001b[43m_message\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mMessage\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43m_CheckCalledFromGeneratedFile\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[0;32m    562\u001b[0m   \u001b[38;5;28;01mif\u001b[39;00m is_extension:\n\u001b[0;32m    563\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m _message\u001b[38;5;241m.\u001b[39mdefault_pool\u001b[38;5;241m.\u001b[39mFindExtensionByName(full_name)\n",
      "\u001b[1;31mTypeError\u001b[0m: Descriptors cannot not be created directly.\nIf this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.\nIf you cannot immediately regenerate your protos, some other possible workarounds are:\n 1. Downgrade the protobuf package to 3.20.x or lower.\n 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).\n\nMore information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "from matplotlib import style\n",
    "import seaborn as sns\n",
    "import random\n",
    "\n",
    "%matplotlib inline\n",
    "style.use('fivethirtyeight')\n",
    "sns.set(style='whitegrid', color_codes=True)\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.metrics import confusion_matrix\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, roc_curve, roc_auc_score\n",
    "from sklearn.model_selection import GridSearchCV\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras import backend as K\n",
    "from keras.models import Sequential\n",
    "from keras.layers import Dense, Dropout, Flatten, Activation\n",
    "from keras.layers import Conv2D, MaxPooling2D, BatchNormalization, GlobalAveragePooling2D\n",
    "from tensorflow.keras.optimizers import Adam, SGD, Adagrad, Adadelta, RMSprop\n",
    "from tensorflow.keras.utils import to_categorical\n",
    "from keras.preprocessing.image import ImageDataGenerator\n",
    "from keras.callbacks import ModelCheckpoint, EarlyStopping, TensorBoard, CSVLogger, ReduceLROnPlateau\n",
    "from keras.models import load_model\n",
    "from keras.applications.inception_v3 import decode_predictions\n",
    "\n",
    "import tensorflow as tf\n",
    "import random as rn\n",
    "\n",
    "import os\n",
    "from random import shuffle\n",
    "from zipfile import ZipFile\n",
    "from PIL import Image"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(20000, 120, 320)\n"
     ]
    }
   ],
   "source": [
    "input_width = 320\n",
    "input_height = 120 \n",
    "n_hid = 1000\n",
    "\n",
    "lookup = dict()\n",
    "reverselookup = dict()\n",
    "count = 0\n",
    "for j in os.listdir('./input/leapgestrecog/leapGestRecog/00/'):\n",
    "    if not j.startswith('.'): # If running this code locally, this is to \n",
    "                              # ensure you aren't reading in hidden folders\n",
    "        lookup[j] = count\n",
    "        reverselookup[count] = j\n",
    "        count = count + 1\n",
    "lookup\n",
    "\n",
    "\n",
    "x_data = []\n",
    "y_data = []\n",
    "datacount = 0 # We'll use this to tally how many images are in our dataset\n",
    "for i in range(0, 10): # Loop over the ten top-level folders\n",
    "    for j in os.listdir('./input/leapgestrecog/leapGestRecog/0' + str(i) + '/'):\n",
    "        if not j.startswith('.'): # Again avoid hidden folders\n",
    "            count = 0 # To tally images of a given gesture\n",
    "            for k in os.listdir('./input/leapgestrecog/leapGestRecog/0' + \n",
    "                                str(i) + '/' + j + '/'):\n",
    "                                # Loop over the images\n",
    "                img = Image.open('./input/leapgestrecog/leapGestRecog/0' + \n",
    "                                 str(i) + '/' + j + '/' + k).convert('L')\n",
    "                                # Read in and convert to greyscale\n",
    "                img = img.resize((input_width, input_height))\n",
    "                arr = np.array(img)\n",
    "                x_data.append(arr) \n",
    "                count = count + 1\n",
    "            y_values = np.full((count, 1), lookup[j]) \n",
    "            y_data.append(y_values)\n",
    "            datacount = datacount + count\n",
    "x_data = np.array(x_data, dtype = 'float32')\n",
    "x_data = x_data / 255 * 2 - 1\n",
    "print(x_data.shape)\n",
    "#x_data = x_data.reshape(datacount, -1)\n",
    "\n",
    "y_data = np.array(y_data)\n",
    "y_data = y_data.reshape(datacount)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(20000, 120, 320, 1)\n",
      "(20000, 10)\n"
     ]
    }
   ],
   "source": [
    "y_data=to_categorical(y_data)\n",
    "x_data = x_data.reshape((datacount, input_height, input_width, 1))\n",
    "\n",
    "print(x_data.shape)\n",
    "print(y_data.shape)\n",
    "x_train,x_further,y_train,y_further = train_test_split(x_data,y_data,test_size = 0.2, random_state=42)\n",
    "x_validate,x_test,y_validate,y_test = train_test_split(x_further,y_further,test_size = 0.5, random_state=42)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "loaded_model_aug = load_model('./my_model.h5',compile=False)\n",
    "loaded_model = load_model('./my_model_wo_pre.h5',compile=False)\n",
    "loaded_model_drop = load_model('./my_model_w_dropout.h5',compile=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\User\\anaconda3\\lib\\site-packages\\keras\\optimizers\\optimizer_v2\\adam.py:114: UserWarning: The `lr` argument is deprecated, use `learning_rate` instead.\n",
      "  super().__init__(name, **kwargs)\n"
     ]
    }
   ],
   "source": [
    "loaded_model.compile(optimizer=Adam(lr=0.001),loss='categorical_crossentropy',metrics=['accuracy'])\n",
    "loaded_model_aug.compile(optimizer=Adam(lr=0.001),loss='categorical_crossentropy',metrics=['accuracy'])\n",
    "loaded_model_drop.compile(optimizer=Adam(lr=0.001),loss='categorical_crossentropy',metrics=['accuracy'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(120, 320, 1)\n"
     ]
    }
   ],
   "source": [
    "print(x_data[0].shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "my_pred = x_data[1].reshape(1,input_height, input_width,1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1/1 [==============================] - 0s 24ms/step\n"
     ]
    }
   ],
   "source": [
    "y_pred = loaded_model_aug.predict(my_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "63/63 [==============================] - 7s 109ms/step\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x800 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "y_pred = loaded_model_drop.predict(x_test)\n",
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "y_true = np.argmax(y_test, axis=1)\n",
    "conf_mat = confusion_matrix(y_true, y_pred_classes)\n",
    "\n",
    "plt.figure(figsize=(10, 8))\n",
    "sns.heatmap(conf_mat, annot=True, fmt='d', cmap='Blues', xticklabels=reverselookup.values(),\n",
    "            yticklabels=reverselookup.values())\n",
    "plt.title('Confusion Matrix')\n",
    "plt.xlabel('Predicted')\n",
    "plt.ylabel('True')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import precision_score, recall_score\n",
    "\n",
    "def f2_score(y_true, y_pred):\n",
    "    # Assuming y_true and y_pred are binary (0 or 1) arrays\n",
    "\n",
    "    # Calculate precision and recall\n",
    "    precision = precision_score(y_true, y_pred, average='macro')\n",
    "    recall = recall_score(y_true, y_pred, average='macro')\n",
    "\n",
    "    # Set beta for F2 score\n",
    "    beta = 2\n",
    "\n",
    "    # Calculate F2 score\n",
    "    f2 = (1 + beta**2) * (precision * recall) / ((beta**2 * precision) + recall)\n",
    "\n",
    "    return f2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(2000, 10)\n"
     ]
    }
   ],
   "source": [
    "print(y_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1.0\n"
     ]
    }
   ],
   "source": [
    "print(f2_score(np.argmax(y_test, axis=1), np.argmax(y_pred, axis=1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred_classes = np.argmax(y_pred, axis=1)\n",
    "y_true = np.argmax(y_test, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Requirement already satisfied: tensorflow in c:\\users\\user\\appdata\\roaming\\python\\python310\\site-packages (2.10.1)\n",
      "Collecting tensorflow\n",
      "  Downloading tensorflow-2.15.0-cp310-cp310-win_amd64.whl (2.1 kB)\n",
      "Collecting tensorflow-intel==2.15.0\n",
      "  Downloading tensorflow_intel-2.15.0-cp310-cp310-win_amd64.whl (300.9 MB)\n",
      "     -------------------------------------- 300.9/300.9 MB 2.4 MB/s eta 0:00:00\n",
      "Requirement already satisfied: ml-dtypes~=0.2.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (0.2.0)\n",
      "Requirement already satisfied: packaging in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (23.2)\n",
      "Requirement already satisfied: astunparse>=1.6.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (1.6.3)\n",
      "Requirement already satisfied: setuptools in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (65.6.3)\n",
      "Requirement already satisfied: grpcio<2.0,>=1.24.3 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (1.54.2)\n",
      "Requirement already satisfied: opt-einsum>=2.3.2 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (3.3.0)\n",
      "Collecting keras<2.16,>=2.15.0\n",
      "  Downloading keras-2.15.0-py3-none-any.whl (1.7 MB)\n",
      "     ---------------------------------------- 1.7/1.7 MB 12.1 MB/s eta 0:00:00\n",
      "Requirement already satisfied: flatbuffers>=23.5.26 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (23.5.26)\n",
      "Collecting protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3\n",
      "  Downloading protobuf-4.25.2-cp310-abi3-win_amd64.whl (413 kB)\n",
      "     -------------------------------------- 413.4/413.4 kB 5.2 MB/s eta 0:00:00\n",
      "Requirement already satisfied: typing-extensions>=3.6.6 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (4.4.0)\n",
      "Requirement already satisfied: wrapt<1.15,>=1.11.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (1.14.1)\n",
      "Requirement already satisfied: six>=1.12.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (1.16.0)\n",
      "Collecting tensorflow-estimator<2.16,>=2.15.0\n",
      "  Downloading tensorflow_estimator-2.15.0-py2.py3-none-any.whl (441 kB)\n",
      "     -------------------------------------- 442.0/442.0 kB 7.0 MB/s eta 0:00:00\n",
      "Requirement already satisfied: absl-py>=1.0.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (1.4.0)\n",
      "Collecting tensorboard<2.16,>=2.15\n",
      "  Downloading tensorboard-2.15.1-py3-none-any.whl (5.5 MB)\n",
      "     ---------------------------------------- 5.5/5.5 MB 11.4 MB/s eta 0:00:00\n",
      "Requirement already satisfied: h5py>=2.9.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (3.7.0)\n",
      "Requirement already satisfied: numpy<2.0.0,>=1.23.5 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (1.23.5)\n",
      "Requirement already satisfied: libclang>=13.0.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (16.0.0)\n",
      "Requirement already satisfied: google-pasta>=0.1.1 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (0.2.0)\n",
      "Requirement already satisfied: gast!=0.5.0,!=0.5.1,!=0.5.2,>=0.2.1 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (0.4.0)\n",
      "Requirement already satisfied: tensorflow-io-gcs-filesystem>=0.23.1 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (0.31.0)\n",
      "Requirement already satisfied: termcolor>=1.1.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorflow-intel==2.15.0->tensorflow) (2.3.0)\n",
      "Requirement already satisfied: wheel<1.0,>=0.23.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from astunparse>=1.6.0->tensorflow-intel==2.15.0->tensorflow) (0.38.4)\n",
      "Requirement already satisfied: markdown>=2.6.8 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (3.4.1)\n",
      "Collecting protobuf!=4.21.0,!=4.21.1,!=4.21.2,!=4.21.3,!=4.21.4,!=4.21.5,<5.0.0dev,>=3.20.3\n",
      "  Downloading protobuf-4.23.4-cp310-abi3-win_amd64.whl (422 kB)\n",
      "     -------------------------------------- 422.5/422.5 kB 8.8 MB/s eta 0:00:00\n",
      "Collecting tensorboard-data-server<0.8.0,>=0.7.0\n",
      "  Downloading tensorboard_data_server-0.7.2-py3-none-any.whl (2.4 kB)\n",
      "Requirement already satisfied: google-auth<3,>=1.6.3 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (2.20.0)\n",
      "Collecting google-auth-oauthlib<2,>=0.5\n",
      "  Downloading google_auth_oauthlib-1.2.0-py2.py3-none-any.whl (24 kB)\n",
      "Requirement already satisfied: werkzeug>=1.0.1 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (2.2.2)\n",
      "Requirement already satisfied: requests<3,>=2.21.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (2.31.0)\n",
      "Requirement already satisfied: pyasn1-modules>=0.2.1 in c:\\users\\user\\anaconda3\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (0.2.8)\n",
      "Requirement already satisfied: urllib3<2.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (1.26.14)\n",
      "Requirement already satisfied: cachetools<6.0,>=2.0.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (5.3.1)\n",
      "Requirement already satisfied: rsa<5,>=3.1.4 in c:\\users\\user\\anaconda3\\lib\\site-packages (from google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (4.9)\n",
      "Requirement already satisfied: requests-oauthlib>=0.7.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (1.3.1)\n",
      "Requirement already satisfied: idna<4,>=2.5 in c:\\users\\user\\anaconda3\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (3.4)\n",
      "Requirement already satisfied: charset-normalizer<4,>=2 in c:\\users\\user\\anaconda3\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (2.0.4)\n",
      "Requirement already satisfied: certifi>=2017.4.17 in c:\\users\\user\\anaconda3\\lib\\site-packages (from requests<3,>=2.21.0->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (2022.12.7)\n",
      "Requirement already satisfied: MarkupSafe>=2.1.1 in c:\\users\\user\\anaconda3\\lib\\site-packages (from werkzeug>=1.0.1->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (2.1.1)\n",
      "Requirement already satisfied: pyasn1<0.5.0,>=0.4.6 in c:\\users\\user\\anaconda3\\lib\\site-packages (from pyasn1-modules>=0.2.1->google-auth<3,>=1.6.3->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (0.4.8)\n",
      "Requirement already satisfied: oauthlib>=3.0.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from requests-oauthlib>=0.7.0->google-auth-oauthlib<2,>=0.5->tensorboard<2.16,>=2.15->tensorflow-intel==2.15.0->tensorflow) (3.2.2)\n",
      "Installing collected packages: tensorflow-estimator, tensorboard-data-server, protobuf, keras, google-auth-oauthlib, tensorboard, tensorflow-intel, tensorflow\n",
      "  Attempting uninstall: tensorflow-estimator\n",
      "    Found existing installation: tensorflow-estimator 2.10.0\n",
      "    Uninstalling tensorflow-estimator-2.10.0:\n",
      "      Successfully uninstalled tensorflow-estimator-2.10.0\n",
      "  Attempting uninstall: tensorboard-data-server\n",
      "    Found existing installation: tensorboard-data-server 0.6.1\n",
      "    Uninstalling tensorboard-data-server-0.6.1:\n",
      "      Successfully uninstalled tensorboard-data-server-0.6.1\n",
      "  Attempting uninstall: protobuf\n",
      "    Found existing installation: protobuf 3.19.6\n",
      "    Uninstalling protobuf-3.19.6:\n",
      "      Successfully uninstalled protobuf-3.19.6\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "    WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "    WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "    WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "ERROR: Could not install packages due to an OSError: [WinError 5] Zugriff verweigert: 'C:\\\\Users\\\\User\\\\anaconda3\\\\Lib\\\\site-packages\\\\google\\\\~-otobuf\\\\internal\\\\_api_implementation.cp310-win_amd64.pyd'\n",
      "Consider using the `--user` option or check the permissions.\n",
      "\n",
      "WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n",
      "WARNING: Ignoring invalid distribution -rotobuf (c:\\users\\user\\anaconda3\\lib\\site-packages)\n"
     ]
    }
   ],
   "source": [
    "%pip install --upgrade tensorflow"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "base",
   "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": "3.10.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}