Datasets
Models
Downloads: 1
Diff of
/notebooks/cancer-detection_from_histology.ipynb
[000000]
..
[291e9d]
Switch to unified view
| a | b/notebooks/cancer-detection_from_histology.ipynb | ||
|---|---|---|---|
| 1 | { |
||
| 2 | "cells": [ |
||
| 3 | { |
||
| 4 | "cell_type": "markdown", |
||
| 5 | "metadata": {}, |
||
| 6 | "source": [ |
||
| 7 | "# 1 \n", |
||
| 8 | "\n", |
||
| 9 | "# a). Problem Statment\n", |
||
| 10 | "\n", |
||
| 11 | "> ### Task - The problem is mainly a BINARY IMAGE CLASSIFICATION PROBLEM. The Problem focuses on identifying the presence of metastases from a 96 * 96 digital histopathology images\n", |
||
| 12 | "\n", |
||
| 13 | "\n", |
||
| 14 | "\n", |
||
| 15 | "\n", |
||
| 16 | "# b). Analysis of the problem Statment\n", |
||
| 17 | "\n", |
||
| 18 | "> ## What Exactly the problem statment conveys to us?\n", |
||
| 19 | "> ### 1. The problem deals with the Binary Classification of the Image that has a shape of 96px * 96px. It involves identifying metastases from the 96px * 96px digital images.\n", |
||
| 20 | "\n" |
||
| 21 | ] |
||
| 22 | }, |
||
| 23 | { |
||
| 24 | "cell_type": "markdown", |
||
| 25 | "metadata": {}, |
||
| 26 | "source": [ |
||
| 27 | "\n", |
||
| 28 | "\n" |
||
| 29 | ] |
||
| 30 | }, |
||
| 31 | { |
||
| 32 | "cell_type": "markdown", |
||
| 33 | "metadata": {}, |
||
| 34 | "source": [ |
||
| 35 | "### So, let's dive into the domain involving Data Collection:\n", |
||
| 36 | "\n", |
||
| 37 | "- **The data provided for classification consists of histopathological images.** These are glass slide microscope images of lymph nodes.\n", |
||
| 38 | "\n", |
||
| 39 | "- **Typically, nuclei are stained blue**, while the cytoplasm and extracellular parts appear in various shades of pink.\n", |
||
| 40 | "\n", |
||
| 41 | "- **Lymph nodes are small glands** that filter the fluid in the lymphatic system, and they are often the first place breast cancer spreads to.\n", |
||
| 42 | "\n", |
||
| 43 | "- **Histological assessment of lymph node metastases** is integral to determining the stage of breast cancer in the TNM classification, a globally recognized standard for classifying cancer spread.\n" |
||
| 44 | ] |
||
| 45 | }, |
||
| 46 | { |
||
| 47 | "cell_type": "markdown", |
||
| 48 | "metadata": {}, |
||
| 49 | "source": [ |
||
| 50 | "# 2. Data Understanding\n", |
||
| 51 | "\n", |
||
| 52 | "* The dataset contains the histopathological Images, each image is 96px * 96px. \n", |
||
| 53 | "\n", |
||
| 54 | "* A positive label indicates that the center 32x32px region of a patch contains at least one pixel of tumor tissue. Tumor tissue in the outer region of the patch does not influence the label. This outer region is provided to enable fully-convolutional models that do not use zero-padding, to ensure consistent behavior when applied to a whole-slide image.\n", |
||
| 55 | "\n", |
||
| 56 | "\n", |
||
| 57 | "\n", |
||
| 58 | "\n", |
||
| 59 | "\n", |
||
| 60 | "\n", |
||
| 61 | "* ### **IS DATA RELEVANT TO THE PROBLEM ?**\n", |
||
| 62 | "> This dataset is a combination of two independent datasets collected in Radboud University Medical Center (Nijmegen, the Netherlands), and the University Medical Center Utrecht (Utrecht, the Netherlands).\n" |
||
| 63 | ] |
||
| 64 | }, |
||
| 65 | { |
||
| 66 | "cell_type": "markdown", |
||
| 67 | "metadata": {}, |
||
| 68 | "source": [ |
||
| 69 | "# 3. Designing the Model (Coding Part)" |
||
| 70 | ] |
||
| 71 | }, |
||
| 72 | { |
||
| 73 | "cell_type": "code", |
||
| 74 | "execution_count": 1, |
||
| 75 | "metadata": { |
||
| 76 | "_cell_guid": "79c7e3d0-c299-4dcb-8224-4455121ee9b0", |
||
| 77 | "_uuid": "d629ff2d2480ee46fbb7e2d37f6b5fab8052498a" |
||
| 78 | }, |
||
| 79 | "outputs": [], |
||
| 80 | "source": [ |
||
| 81 | "# Importing Libraries\n", |
||
| 82 | "from numpy.random import seed\n", |
||
| 83 | "seed(101)\n", |
||
| 84 | "\n", |
||
| 85 | "import pandas as pd\n", |
||
| 86 | "import numpy as np\n", |
||
| 87 | "\n", |
||
| 88 | "\n", |
||
| 89 | "import tensorflow as tf\n", |
||
| 90 | "from tensorflow import keras\n", |
||
| 91 | "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n", |
||
| 92 | "from tensorflow.keras.layers import Conv2D, MaxPooling2D\n", |
||
| 93 | "from tensorflow.keras.layers import Dense, Dropout, Flatten, Activation\n", |
||
| 94 | "from tensorflow.keras.models import Sequential\n", |
||
| 95 | "from tensorflow.keras.callbacks import EarlyStopping, ReduceLROnPlateau, ModelCheckpoint\n", |
||
| 96 | "from tensorflow.keras.optimizers import Adam\n", |
||
| 97 | "\n", |
||
| 98 | "import os\n", |
||
| 99 | "import cv2\n", |
||
| 100 | "\n", |
||
| 101 | "from sklearn.utils import shuffle\n", |
||
| 102 | "from sklearn.metrics import confusion_matrix\n", |
||
| 103 | "from sklearn.model_selection import train_test_split\n", |
||
| 104 | "import itertools\n", |
||
| 105 | "import shutil\n", |
||
| 106 | "import matplotlib.pyplot as plt\n", |
||
| 107 | "%matplotlib inline\n", |
||
| 108 | "tf.random.set_seed(101)" |
||
| 109 | ] |
||
| 110 | }, |
||
| 111 | { |
||
| 112 | "cell_type": "code", |
||
| 113 | "execution_count": 2, |
||
| 114 | "metadata": {}, |
||
| 115 | "outputs": [], |
||
| 116 | "source": [ |
||
| 117 | "# Setting Some Pre-Requisites\n", |
||
| 118 | "IMAGE_SIZE=96\n", |
||
| 119 | "IMAGE_CHANNELS=3\n", |
||
| 120 | "SAMPLE_SIZE=80000 # We will be training 80,000 samples from each label" |
||
| 121 | ] |
||
| 122 | }, |
||
| 123 | { |
||
| 124 | "cell_type": "code", |
||
| 125 | "execution_count": 4, |
||
| 126 | "metadata": {}, |
||
| 127 | "outputs": [ |
||
| 128 | { |
||
| 129 | "data": { |
||
| 130 | "text/plain": [ |
||
| 131 | "['train_labels.csv', 'test', 'train', 'sample_submission.csv']" |
||
| 132 | ] |
||
| 133 | }, |
||
| 134 | "execution_count": 4, |
||
| 135 | "metadata": {}, |
||
| 136 | "output_type": "execute_result" |
||
| 137 | } |
||
| 138 | ], |
||
| 139 | "source": [ |
||
| 140 | "# So, what are the files which are available?\n", |
||
| 141 | "\n", |
||
| 142 | "os.listdir('./histopathologic-cancer-detection')" |
||
| 143 | ] |
||
| 144 | }, |
||
| 145 | { |
||
| 146 | "cell_type": "code", |
||
| 147 | "execution_count": 6, |
||
| 148 | "metadata": {}, |
||
| 149 | "outputs": [ |
||
| 150 | { |
||
| 151 | "name": "stdout", |
||
| 152 | "output_type": "stream", |
||
| 153 | "text": [ |
||
| 154 | "220025\n", |
||
| 155 | "57458\n" |
||
| 156 | ] |
||
| 157 | } |
||
| 158 | ], |
||
| 159 | "source": [ |
||
| 160 | "# So, how many images are there in each of the folder in the training dataset?\n", |
||
| 161 | "\n", |
||
| 162 | "print(len(os.listdir('./histopathologic-cancer-detection/train')))\n", |
||
| 163 | "print(len(os.listdir('./histopathologic-cancer-detection/test')))" |
||
| 164 | ] |
||
| 165 | }, |
||
| 166 | { |
||
| 167 | "cell_type": "code", |
||
| 168 | "execution_count": 8, |
||
| 169 | "metadata": {}, |
||
| 170 | "outputs": [ |
||
| 171 | { |
||
| 172 | "name": "stdout", |
||
| 173 | "output_type": "stream", |
||
| 174 | "text": [ |
||
| 175 | "(220025, 2)\n" |
||
| 176 | ] |
||
| 177 | } |
||
| 178 | ], |
||
| 179 | "source": [ |
||
| 180 | "# Creating a dataframe of all the training images\n", |
||
| 181 | "\n", |
||
| 182 | "df_data = pd.read_csv('./histopathologic-cancer-detection/train_labels.csv')\n", |
||
| 183 | "\n", |
||
| 184 | "# removing this image because it caused a training error previously\n", |
||
| 185 | "df_data[df_data['id'] != 'dd6dfed324f9fcb6f93f46f32fc800f2ec196be2']\n", |
||
| 186 | "\n", |
||
| 187 | "# removing this image because it's black\n", |
||
| 188 | "df_data[df_data['id'] != '9369c7278ec8bcc6c880d99194de09fc2bd4efbe']\n", |
||
| 189 | "\n", |
||
| 190 | "\n", |
||
| 191 | "print(df_data.shape)" |
||
| 192 | ] |
||
| 193 | }, |
||
| 194 | { |
||
| 195 | "cell_type": "code", |
||
| 196 | "execution_count": 9, |
||
| 197 | "metadata": {}, |
||
| 198 | "outputs": [ |
||
| 199 | { |
||
| 200 | "data": { |
||
| 201 | "text/plain": [ |
||
| 202 | "0 130908\n", |
||
| 203 | "1 89117\n", |
||
| 204 | "Name: label, dtype: int64" |
||
| 205 | ] |
||
| 206 | }, |
||
| 207 | "execution_count": 9, |
||
| 208 | "metadata": {}, |
||
| 209 | "output_type": "execute_result" |
||
| 210 | } |
||
| 211 | ], |
||
| 212 | "source": [ |
||
| 213 | "df_data['label'].value_counts()" |
||
| 214 | ] |
||
| 215 | }, |
||
| 216 | { |
||
| 217 | "cell_type": "code", |
||
| 218 | "execution_count": 10, |
||
| 219 | "metadata": {}, |
||
| 220 | "outputs": [], |
||
| 221 | "source": [ |
||
| 222 | "# source: https://www.kaggle.com/gpreda/honey-bee-subspecies-classification\n", |
||
| 223 | "\n", |
||
| 224 | "def draw_category_images(col_name,figure_cols, df, IMAGE_PATH):\n", |
||
| 225 | " categories = df[col_name].unique()\n", |
||
| 226 | " num_rows = len(categories)\n", |
||
| 227 | " \n", |
||
| 228 | " fig, axes = plt.subplots(nrows=num_rows, ncols=num_cols, figsize=(4 * num_cols, 4 * num_rows))\n", |
||
| 229 | "\n", |
||
| 230 | " for i, category in enumerate(categories):\n", |
||
| 231 | " sample_df = df[df[col_name] == category].sample(num_cols)\n", |
||
| 232 | " \n", |
||
| 233 | " for j, (_, sample_row) in enumerate(sample_df.iterrows()):\n", |
||
| 234 | " image_file_path = f\"{image_dir}{sample_row['id']}.tif\"\n", |
||
| 235 | " image = cv2.imread(image_file_path)\n", |
||
| 236 | " \n", |
||
| 237 | " axes[i, j].imshow(image, resample=True, cmap='gray')\n", |
||
| 238 | " axes[i, j].set_title(category, fontsize=16)\n", |
||
| 239 | "\n", |
||
| 240 | " plt.tight_layout()\n", |
||
| 241 | " plt.show()\n", |
||
| 242 | "\n", |
||
| 243 | "\n", |
||
| 244 | "\n", |
||
| 245 | "\n", |
||
| 246 | "\n", |
||
| 247 | "\n" |
||
| 248 | ] |
||
| 249 | }, |
||
| 250 | { |
||
| 251 | "cell_type": "code", |
||
| 252 | "execution_count": 12, |
||
| 253 | "metadata": {}, |
||
| 254 | "outputs": [ |
||
| 255 | { |
||
| 256 | "data": { |
||
| 257 | "image/png": "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\n", |
||
| 258 | "text/plain": [ |
||
| 259 | "<Figure size 1600x800 with 8 Axes>" |
||
| 260 | ] |
||
| 261 | }, |
||
| 262 | "metadata": {}, |
||
| 263 | "output_type": "display_data" |
||
| 264 | } |
||
| 265 | ], |
||
| 266 | "source": [ |
||
| 267 | "IMAGE_PATH = './histopathologic-cancer-detection/train/' \n", |
||
| 268 | "\n", |
||
| 269 | "draw_category_images('label',4, df_data, IMAGE_PATH)" |
||
| 270 | ] |
||
| 271 | }, |
||
| 272 | { |
||
| 273 | "cell_type": "code", |
||
| 274 | "execution_count": 13, |
||
| 275 | "metadata": {}, |
||
| 276 | "outputs": [ |
||
| 277 | { |
||
| 278 | "data": { |
||
| 279 | "text/plain": [ |
||
| 280 | "0 80000\n", |
||
| 281 | "1 80000\n", |
||
| 282 | "Name: label, dtype: int64" |
||
| 283 | ] |
||
| 284 | }, |
||
| 285 | "execution_count": 13, |
||
| 286 | "metadata": {}, |
||
| 287 | "output_type": "execute_result" |
||
| 288 | } |
||
| 289 | ], |
||
| 290 | "source": [ |
||
| 291 | "# Sample data based on label\n", |
||
| 292 | "df_0 = df_data[df_data['label'] == 0].sample(SAMPLE_SIZE, random_state=42)\n", |
||
| 293 | "df_1 = df_data[df_data['label'] == 1].sample(SAMPLE_SIZE, random_state=42)\n", |
||
| 294 | "\n", |
||
| 295 | "# Combine and shuffle the sampled data\n", |
||
| 296 | "df_data = pd.concat([df_0, df_1], axis=0).reset_index(drop=True).sample(frac=1).reset_index(drop=True)\n", |
||
| 297 | "\n", |
||
| 298 | "# Display label counts\n", |
||
| 299 | "df_data['label'].value_counts()" |
||
| 300 | ] |
||
| 301 | }, |
||
| 302 | { |
||
| 303 | "cell_type": "code", |
||
| 304 | "execution_count": 14, |
||
| 305 | "metadata": {}, |
||
| 306 | "outputs": [ |
||
| 307 | { |
||
| 308 | "name": "stdout", |
||
| 309 | "output_type": "stream", |
||
| 310 | "text": [ |
||
| 311 | "(144000, 2)\n", |
||
| 312 | "(16000, 2)\n" |
||
| 313 | ] |
||
| 314 | } |
||
| 315 | ], |
||
| 316 | "source": [ |
||
| 317 | "# Now, for the train-test split\n", |
||
| 318 | "\n", |
||
| 319 | "# stratify=y creates a balanced validation set.\n", |
||
| 320 | "y = df_data['label']\n", |
||
| 321 | "\n", |
||
| 322 | "df_train, df_val = train_test_split(df_data, test_size=0.10, random_state=101, stratify=y)\n", |
||
| 323 | "\n", |
||
| 324 | "print(df_train.shape)\n", |
||
| 325 | "print(df_val.shape)" |
||
| 326 | ] |
||
| 327 | }, |
||
| 328 | { |
||
| 329 | "cell_type": "code", |
||
| 330 | "execution_count": 15, |
||
| 331 | "metadata": {}, |
||
| 332 | "outputs": [], |
||
| 333 | "source": [ |
||
| 334 | "# Create a new directory so that we will be using the ImageDataGenerator\n", |
||
| 335 | "base_dir = 'base_dir'\n", |
||
| 336 | "os.makedirs(base_dir)\n", |
||
| 337 | "\n", |
||
| 338 | "sub_dirs = ['train_dir', 'val_dir']\n", |
||
| 339 | "classes = ['a_no_tumor_tissue', 'b_has_tumor_tissue']\n", |
||
| 340 | "\n", |
||
| 341 | "# Create train and validation directories with class sub-directories\n", |
||
| 342 | "for sub_dir in sub_dirs:\n", |
||
| 343 | " for class_name in classes:\n", |
||
| 344 | " os.makedirs(os.path.join(base_dir, sub_dir, class_name))\n", |
||
| 345 | "# check that the folders have been created\n", |
||
| 346 | "os.listdir('base_dir/train_dir')" |
||
| 347 | ] |
||
| 348 | }, |
||
| 349 | { |
||
| 350 | "cell_type": "code", |
||
| 351 | "execution_count": 16, |
||
| 352 | "metadata": {}, |
||
| 353 | "outputs": [ |
||
| 354 | { |
||
| 355 | "data": { |
||
| 356 | "text/plain": [ |
||
| 357 | "['a_no_tumor_tissue', 'b_has_tumor_tissue']" |
||
| 358 | ] |
||
| 359 | }, |
||
| 360 | "execution_count": 16, |
||
| 361 | "metadata": {}, |
||
| 362 | "output_type": "execute_result" |
||
| 363 | } |
||
| 364 | ], |
||
| 365 | "source": [ |
||
| 366 | "# check that the folders have been created\n", |
||
| 367 | "os.listdir('base_dir/train_dir')" |
||
| 368 | ] |
||
| 369 | }, |
||
| 370 | { |
||
| 371 | "cell_type": "code", |
||
| 372 | "execution_count": 17, |
||
| 373 | "metadata": {}, |
||
| 374 | "outputs": [], |
||
| 375 | "source": [ |
||
| 376 | "# Set the id as the index in df_data\n", |
||
| 377 | "df_data.set_index('id', inplace=True)" |
||
| 378 | ] |
||
| 379 | }, |
||
| 380 | { |
||
| 381 | "cell_type": "code", |
||
| 382 | "execution_count": 20, |
||
| 383 | "metadata": {}, |
||
| 384 | "outputs": [], |
||
| 385 | "source": [ |
||
| 386 | "import os\n", |
||
| 387 | "import shutil\n", |
||
| 388 | "\n", |
||
| 389 | "def transfer_images(image_list, destination_dir, df):\n", |
||
| 390 | " for image in image_list:\n", |
||
| 391 | " fname = f\"{image}.tif\"\n", |
||
| 392 | " target = df.loc[image, 'label']\n", |
||
| 393 | " \n", |
||
| 394 | " label = 'a_no_tumor_tissue' if target == 0 else 'b_has_tumor_tissue'\n", |
||
| 395 | "\n", |
||
| 396 | " src = os.path.join('./histopathologic-cancer-detection/train', fname)\n", |
||
| 397 | " dst = os.path.join(destination_dir, label, fname)\n", |
||
| 398 | " \n", |
||
| 399 | " shutil.copyfile(src, dst)\n", |
||
| 400 | "\n", |
||
| 401 | "# Get a list of train and val images\n", |
||
| 402 | "train_list = list(df_train['id'])\n", |
||
| 403 | "val_list = list(df_val['id'])\n", |
||
| 404 | "\n", |
||
| 405 | "# Transfer the train and val images\n", |
||
| 406 | "transfer_images(train_list, train_dir, df_data)\n", |
||
| 407 | "transfer_images(val_list, val_dir, df_data)" |
||
| 408 | ] |
||
| 409 | }, |
||
| 410 | { |
||
| 411 | "cell_type": "code", |
||
| 412 | "execution_count": 21, |
||
| 413 | "metadata": {}, |
||
| 414 | "outputs": [ |
||
| 415 | { |
||
| 416 | "name": "stdout", |
||
| 417 | "output_type": "stream", |
||
| 418 | "text": [ |
||
| 419 | "72000\n", |
||
| 420 | "72000\n" |
||
| 421 | ] |
||
| 422 | } |
||
| 423 | ], |
||
| 424 | "source": [ |
||
| 425 | "# check how many train images we have in each folder\n", |
||
| 426 | "\n", |
||
| 427 | "print(len(os.listdir('base_dir/train_dir/a_no_tumor_tissue')))\n", |
||
| 428 | "print(len(os.listdir('base_dir/train_dir/b_has_tumor_tissue')))" |
||
| 429 | ] |
||
| 430 | }, |
||
| 431 | { |
||
| 432 | "cell_type": "code", |
||
| 433 | "execution_count": 22, |
||
| 434 | "metadata": {}, |
||
| 435 | "outputs": [ |
||
| 436 | { |
||
| 437 | "name": "stdout", |
||
| 438 | "output_type": "stream", |
||
| 439 | "text": [ |
||
| 440 | "8000\n", |
||
| 441 | "8000\n" |
||
| 442 | ] |
||
| 443 | } |
||
| 444 | ], |
||
| 445 | "source": [ |
||
| 446 | "# check how many val images we have in each folder\n", |
||
| 447 | "\n", |
||
| 448 | "print(len(os.listdir('base_dir/val_dir/a_no_tumor_tissue')))\n", |
||
| 449 | "print(len(os.listdir('base_dir/val_dir/b_has_tumor_tissue')))" |
||
| 450 | ] |
||
| 451 | }, |
||
| 452 | { |
||
| 453 | "cell_type": "code", |
||
| 454 | "execution_count": 23, |
||
| 455 | "metadata": {}, |
||
| 456 | "outputs": [], |
||
| 457 | "source": [ |
||
| 458 | "# Set up the generators\n", |
||
| 459 | "train_path = 'base_dir/train_dir'\n", |
||
| 460 | "valid_path = 'base_dir/val_dir'\n", |
||
| 461 | "test_path = './histopathologic-cancer-detection/test'\n", |
||
| 462 | "\n", |
||
| 463 | "num_train_samples = len(df_train)\n", |
||
| 464 | "num_val_samples = len(df_val)\n", |
||
| 465 | "train_batch_size = 10\n", |
||
| 466 | "val_batch_size = 10\n", |
||
| 467 | "\n", |
||
| 468 | "\n", |
||
| 469 | "train_steps = np.ceil(num_train_samples / train_batch_size)\n", |
||
| 470 | "val_steps = np.ceil(num_val_samples / val_batch_size)" |
||
| 471 | ] |
||
| 472 | }, |
||
| 473 | { |
||
| 474 | "cell_type": "code", |
||
| 475 | "execution_count": 24, |
||
| 476 | "metadata": {}, |
||
| 477 | "outputs": [ |
||
| 478 | { |
||
| 479 | "name": "stdout", |
||
| 480 | "output_type": "stream", |
||
| 481 | "text": [ |
||
| 482 | "Found 144000 images belonging to 2 classes.\n", |
||
| 483 | "Found 16000 images belonging to 2 classes.\n", |
||
| 484 | "Found 16000 images belonging to 2 classes.\n" |
||
| 485 | ] |
||
| 486 | } |
||
| 487 | ], |
||
| 488 | "source": [ |
||
| 489 | "datagen = ImageDataGenerator(rescale=1.0/255)\n", |
||
| 490 | "\n", |
||
| 491 | "train_gen = datagen.flow_from_directory(train_path,\n", |
||
| 492 | " target_size=(IMAGE_SIZE,IMAGE_SIZE),\n", |
||
| 493 | " batch_size=train_batch_size,\n", |
||
| 494 | " class_mode='categorical')\n", |
||
| 495 | "\n", |
||
| 496 | "val_gen = datagen.flow_from_directory(valid_path,\n", |
||
| 497 | " target_size=(IMAGE_SIZE,IMAGE_SIZE),\n", |
||
| 498 | " batch_size=val_batch_size,\n", |
||
| 499 | " class_mode='categorical')\n", |
||
| 500 | "\n", |
||
| 501 | "# Note: shuffle=False causes the test dataset to not be shuffled\n", |
||
| 502 | "test_gen = datagen.flow_from_directory(valid_path,\n", |
||
| 503 | " target_size=(IMAGE_SIZE,IMAGE_SIZE),\n", |
||
| 504 | " batch_size=1,\n", |
||
| 505 | " class_mode='categorical',\n", |
||
| 506 | " shuffle=False)" |
||
| 507 | ] |
||
| 508 | }, |
||
| 509 | { |
||
| 510 | "cell_type": "markdown", |
||
| 511 | "metadata": {}, |
||
| 512 | "source": [ |
||
| 513 | "### The model that I have choosen for this problem has been taken from <a href = 'https://www.kaggle.com/fmarazzi/baseline-keras-cnn-roc-fast-10min-0-925-lb'>Baseline Keras CNN</a>" |
||
| 514 | ] |
||
| 515 | }, |
||
| 516 | { |
||
| 517 | "cell_type": "code", |
||
| 518 | "execution_count": 25, |
||
| 519 | "metadata": {}, |
||
| 520 | "outputs": [ |
||
| 521 | { |
||
| 522 | "name": "stdout", |
||
| 523 | "output_type": "stream", |
||
| 524 | "text": [ |
||
| 525 | "Model: \"sequential\"\n", |
||
| 526 | "_________________________________________________________________\n", |
||
| 527 | " Layer (type) Output Shape Param # \n", |
||
| 528 | "=================================================================\n", |
||
| 529 | " conv2d (Conv2D) (None, 94, 94, 32) 896 \n", |
||
| 530 | " \n", |
||
| 531 | " conv2d_1 (Conv2D) (None, 92, 92, 32) 9248 \n", |
||
| 532 | " \n", |
||
| 533 | " conv2d_2 (Conv2D) (None, 90, 90, 32) 9248 \n", |
||
| 534 | " \n", |
||
| 535 | " max_pooling2d (MaxPooling2 (None, 45, 45, 32) 0 \n", |
||
| 536 | " D) \n", |
||
| 537 | " \n", |
||
| 538 | " dropout (Dropout) (None, 45, 45, 32) 0 \n", |
||
| 539 | " \n", |
||
| 540 | " conv2d_3 (Conv2D) (None, 43, 43, 64) 18496 \n", |
||
| 541 | " \n", |
||
| 542 | " conv2d_4 (Conv2D) (None, 41, 41, 64) 36928 \n", |
||
| 543 | " \n", |
||
| 544 | " conv2d_5 (Conv2D) (None, 39, 39, 64) 36928 \n", |
||
| 545 | " \n", |
||
| 546 | " max_pooling2d_1 (MaxPoolin (None, 19, 19, 64) 0 \n", |
||
| 547 | " g2D) \n", |
||
| 548 | " \n", |
||
| 549 | " dropout_1 (Dropout) (None, 19, 19, 64) 0 \n", |
||
| 550 | " \n", |
||
| 551 | " conv2d_6 (Conv2D) (None, 17, 17, 128) 73856 \n", |
||
| 552 | " \n", |
||
| 553 | " conv2d_7 (Conv2D) (None, 15, 15, 128) 147584 \n", |
||
| 554 | " \n", |
||
| 555 | " conv2d_8 (Conv2D) (None, 13, 13, 128) 147584 \n", |
||
| 556 | " \n", |
||
| 557 | " max_pooling2d_2 (MaxPoolin (None, 6, 6, 128) 0 \n", |
||
| 558 | " g2D) \n", |
||
| 559 | " \n", |
||
| 560 | " dropout_2 (Dropout) (None, 6, 6, 128) 0 \n", |
||
| 561 | " \n", |
||
| 562 | " flatten (Flatten) (None, 4608) 0 \n", |
||
| 563 | " \n", |
||
| 564 | " dense (Dense) (None, 256) 1179904 \n", |
||
| 565 | " \n", |
||
| 566 | " dropout_3 (Dropout) (None, 256) 0 \n", |
||
| 567 | " \n", |
||
| 568 | " dense_1 (Dense) (None, 2) 514 \n", |
||
| 569 | " \n", |
||
| 570 | "=================================================================\n", |
||
| 571 | "Total params: 1661186 (6.34 MB)\n", |
||
| 572 | "Trainable params: 1661186 (6.34 MB)\n", |
||
| 573 | "Non-trainable params: 0 (0.00 Byte)\n", |
||
| 574 | "_________________________________________________________________\n" |
||
| 575 | ] |
||
| 576 | } |
||
| 577 | ], |
||
| 578 | "source": [ |
||
| 579 | "kernel_size = (3,3)\n", |
||
| 580 | "pool_size= (2,2)\n", |
||
| 581 | "first_filters = 32\n", |
||
| 582 | "second_filters = 64\n", |
||
| 583 | "third_filters = 128\n", |
||
| 584 | "\n", |
||
| 585 | "dropout_conv = 0.3\n", |
||
| 586 | "dropout_dense = 0.3\n", |
||
| 587 | "\n", |
||
| 588 | "\n", |
||
| 589 | "model = Sequential()\n", |
||
| 590 | "model.add(Conv2D(first_filters, kernel_size, activation = 'relu', input_shape = (96, 96, 3)))\n", |
||
| 591 | "model.add(Conv2D(first_filters, kernel_size, activation = 'relu'))\n", |
||
| 592 | "model.add(Conv2D(first_filters, kernel_size, activation = 'relu'))\n", |
||
| 593 | "model.add(MaxPooling2D(pool_size = pool_size)) \n", |
||
| 594 | "model.add(Dropout(dropout_conv))\n", |
||
| 595 | "\n", |
||
| 596 | "model.add(Conv2D(second_filters, kernel_size, activation ='relu'))\n", |
||
| 597 | "model.add(Conv2D(second_filters, kernel_size, activation ='relu'))\n", |
||
| 598 | "model.add(Conv2D(second_filters, kernel_size, activation ='relu'))\n", |
||
| 599 | "model.add(MaxPooling2D(pool_size = pool_size))\n", |
||
| 600 | "model.add(Dropout(dropout_conv))\n", |
||
| 601 | "\n", |
||
| 602 | "model.add(Conv2D(third_filters, kernel_size, activation ='relu'))\n", |
||
| 603 | "model.add(Conv2D(third_filters, kernel_size, activation ='relu'))\n", |
||
| 604 | "model.add(Conv2D(third_filters, kernel_size, activation ='relu'))\n", |
||
| 605 | "model.add(MaxPooling2D(pool_size = pool_size))\n", |
||
| 606 | "model.add(Dropout(dropout_conv))\n", |
||
| 607 | "\n", |
||
| 608 | "model.add(Flatten())\n", |
||
| 609 | "model.add(Dense(256, activation = \"relu\"))\n", |
||
| 610 | "model.add(Dropout(dropout_dense))\n", |
||
| 611 | "model.add(Dense(2, activation = \"softmax\"))\n", |
||
| 612 | "\n", |
||
| 613 | "model.summary()" |
||
| 614 | ] |
||
| 615 | }, |
||
| 616 | { |
||
| 617 | "cell_type": "code", |
||
| 618 | "execution_count": 26, |
||
| 619 | "metadata": {}, |
||
| 620 | "outputs": [ |
||
| 621 | { |
||
| 622 | "name": "stderr", |
||
| 623 | "output_type": "stream", |
||
| 624 | "text": [ |
||
| 625 | "WARNING:absl:At this time, the v2.11+ optimizer `tf.keras.optimizers.Adam` runs slowly on M1/M2 Macs, please use the legacy Keras optimizer instead, located at `tf.keras.optimizers.legacy.Adam`.\n", |
||
| 626 | "WARNING:absl:`lr` is deprecated in Keras optimizer, please use `learning_rate` or use the legacy optimizer, e.g.,tf.keras.optimizers.legacy.Adam.\n", |
||
| 627 | "WARNING:absl:There is a known slowdown when using v2.11+ Keras optimizers on M1/M2 Macs. Falling back to the legacy Keras optimizer, i.e., `tf.keras.optimizers.legacy.Adam`.\n" |
||
| 628 | ] |
||
| 629 | } |
||
| 630 | ], |
||
| 631 | "source": [ |
||
| 632 | "model.compile(Adam(lr=0.0001), loss='binary_crossentropy', \n", |
||
| 633 | " metrics=['accuracy'])" |
||
| 634 | ] |
||
| 635 | }, |
||
| 636 | { |
||
| 637 | "cell_type": "code", |
||
| 638 | "execution_count": 27, |
||
| 639 | "metadata": {}, |
||
| 640 | "outputs": [ |
||
| 641 | { |
||
| 642 | "name": "stdout", |
||
| 643 | "output_type": "stream", |
||
| 644 | "text": [ |
||
| 645 | "{'a_no_tumor_tissue': 0, 'b_has_tumor_tissue': 1}\n" |
||
| 646 | ] |
||
| 647 | } |
||
| 648 | ], |
||
| 649 | "source": [ |
||
| 650 | "# Get the labels that are associated with each index\n", |
||
| 651 | "print(val_gen.class_indices)" |
||
| 652 | ] |
||
| 653 | }, |
||
| 654 | { |
||
| 655 | "cell_type": "code", |
||
| 656 | "execution_count": 29, |
||
| 657 | "metadata": {}, |
||
| 658 | "outputs": [], |
||
| 659 | "source": [ |
||
| 660 | "train_steps = 500" |
||
| 661 | ] |
||
| 662 | }, |
||
| 663 | { |
||
| 664 | "cell_type": "code", |
||
| 665 | "execution_count": 30, |
||
| 666 | "metadata": {}, |
||
| 667 | "outputs": [ |
||
| 668 | { |
||
| 669 | "name": "stdout", |
||
| 670 | "output_type": "stream", |
||
| 671 | "text": [ |
||
| 672 | "Epoch 1/20\n" |
||
| 673 | ] |
||
| 674 | }, |
||
| 675 | { |
||
| 676 | "name": "stderr", |
||
| 677 | "output_type": "stream", |
||
| 678 | "text": [ |
||
| 679 | "/var/folders/3k/vhmk7kx14ybb600l38fbqg780000gn/T/ipykernel_68365/4098454996.py:11: UserWarning: `Model.fit_generator` is deprecated and will be removed in a future version. Please use `Model.fit`, which supports generators.\n", |
||
| 680 | " history = model.fit_generator(train_gen, steps_per_epoch=train_steps,\n" |
||
| 681 | ] |
||
| 682 | }, |
||
| 683 | { |
||
| 684 | "name": "stdout", |
||
| 685 | "output_type": "stream", |
||
| 686 | "text": [ |
||
| 687 | "500/500 [==============================] - ETA: 0s - loss: 0.6933 - accuracy: 0.4996WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 688 | ] |
||
| 689 | }, |
||
| 690 | { |
||
| 691 | "name": "stderr", |
||
| 692 | "output_type": "stream", |
||
| 693 | "text": [ |
||
| 694 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 695 | ] |
||
| 696 | }, |
||
| 697 | { |
||
| 698 | "name": "stdout", |
||
| 699 | "output_type": "stream", |
||
| 700 | "text": [ |
||
| 701 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 702 | ] |
||
| 703 | }, |
||
| 704 | { |
||
| 705 | "name": "stderr", |
||
| 706 | "output_type": "stream", |
||
| 707 | "text": [ |
||
| 708 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 709 | ] |
||
| 710 | }, |
||
| 711 | { |
||
| 712 | "name": "stdout", |
||
| 713 | "output_type": "stream", |
||
| 714 | "text": [ |
||
| 715 | "500/500 [==============================] - 95s 189ms/step - loss: 0.6933 - accuracy: 0.4996 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 716 | "Epoch 2/20\n", |
||
| 717 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.5030WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 718 | ] |
||
| 719 | }, |
||
| 720 | { |
||
| 721 | "name": "stderr", |
||
| 722 | "output_type": "stream", |
||
| 723 | "text": [ |
||
| 724 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 725 | ] |
||
| 726 | }, |
||
| 727 | { |
||
| 728 | "name": "stdout", |
||
| 729 | "output_type": "stream", |
||
| 730 | "text": [ |
||
| 731 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 732 | ] |
||
| 733 | }, |
||
| 734 | { |
||
| 735 | "name": "stderr", |
||
| 736 | "output_type": "stream", |
||
| 737 | "text": [ |
||
| 738 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 739 | ] |
||
| 740 | }, |
||
| 741 | { |
||
| 742 | "name": "stdout", |
||
| 743 | "output_type": "stream", |
||
| 744 | "text": [ |
||
| 745 | "500/500 [==============================] - 91s 181ms/step - loss: 0.6932 - accuracy: 0.5030 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 746 | "Epoch 3/20\n", |
||
| 747 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.4976WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 748 | ] |
||
| 749 | }, |
||
| 750 | { |
||
| 751 | "name": "stderr", |
||
| 752 | "output_type": "stream", |
||
| 753 | "text": [ |
||
| 754 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 755 | ] |
||
| 756 | }, |
||
| 757 | { |
||
| 758 | "name": "stdout", |
||
| 759 | "output_type": "stream", |
||
| 760 | "text": [ |
||
| 761 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 762 | ] |
||
| 763 | }, |
||
| 764 | { |
||
| 765 | "name": "stderr", |
||
| 766 | "output_type": "stream", |
||
| 767 | "text": [ |
||
| 768 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 769 | ] |
||
| 770 | }, |
||
| 771 | { |
||
| 772 | "name": "stdout", |
||
| 773 | "output_type": "stream", |
||
| 774 | "text": [ |
||
| 775 | "500/500 [==============================] - 83s 167ms/step - loss: 0.6932 - accuracy: 0.4976 - val_loss: 0.6931 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 776 | "Epoch 4/20\n", |
||
| 777 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.4976WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 778 | ] |
||
| 779 | }, |
||
| 780 | { |
||
| 781 | "name": "stderr", |
||
| 782 | "output_type": "stream", |
||
| 783 | "text": [ |
||
| 784 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 785 | ] |
||
| 786 | }, |
||
| 787 | { |
||
| 788 | "name": "stdout", |
||
| 789 | "output_type": "stream", |
||
| 790 | "text": [ |
||
| 791 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 792 | ] |
||
| 793 | }, |
||
| 794 | { |
||
| 795 | "name": "stderr", |
||
| 796 | "output_type": "stream", |
||
| 797 | "text": [ |
||
| 798 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 799 | ] |
||
| 800 | }, |
||
| 801 | { |
||
| 802 | "name": "stdout", |
||
| 803 | "output_type": "stream", |
||
| 804 | "text": [ |
||
| 805 | "500/500 [==============================] - 81s 163ms/step - loss: 0.6932 - accuracy: 0.4976 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 806 | "Epoch 5/20\n", |
||
| 807 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.4976WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 808 | ] |
||
| 809 | }, |
||
| 810 | { |
||
| 811 | "name": "stderr", |
||
| 812 | "output_type": "stream", |
||
| 813 | "text": [ |
||
| 814 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 815 | ] |
||
| 816 | }, |
||
| 817 | { |
||
| 818 | "name": "stdout", |
||
| 819 | "output_type": "stream", |
||
| 820 | "text": [ |
||
| 821 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 822 | ] |
||
| 823 | }, |
||
| 824 | { |
||
| 825 | "name": "stderr", |
||
| 826 | "output_type": "stream", |
||
| 827 | "text": [ |
||
| 828 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 829 | ] |
||
| 830 | }, |
||
| 831 | { |
||
| 832 | "name": "stdout", |
||
| 833 | "output_type": "stream", |
||
| 834 | "text": [ |
||
| 835 | "500/500 [==============================] - 82s 164ms/step - loss: 0.6932 - accuracy: 0.4976 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 836 | "Epoch 6/20\n", |
||
| 837 | "500/500 [==============================] - ETA: 0s - loss: 0.6931 - accuracy: 0.5040WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 838 | ] |
||
| 839 | }, |
||
| 840 | { |
||
| 841 | "name": "stderr", |
||
| 842 | "output_type": "stream", |
||
| 843 | "text": [ |
||
| 844 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 845 | ] |
||
| 846 | }, |
||
| 847 | { |
||
| 848 | "name": "stdout", |
||
| 849 | "output_type": "stream", |
||
| 850 | "text": [ |
||
| 851 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 852 | ] |
||
| 853 | }, |
||
| 854 | { |
||
| 855 | "name": "stderr", |
||
| 856 | "output_type": "stream", |
||
| 857 | "text": [ |
||
| 858 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 859 | ] |
||
| 860 | }, |
||
| 861 | { |
||
| 862 | "name": "stdout", |
||
| 863 | "output_type": "stream", |
||
| 864 | "text": [ |
||
| 865 | "500/500 [==============================] - 83s 166ms/step - loss: 0.6931 - accuracy: 0.5040 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 866 | "Epoch 7/20\n", |
||
| 867 | "500/500 [==============================] - ETA: 0s - loss: 0.6931 - accuracy: 0.5096WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 868 | ] |
||
| 869 | }, |
||
| 870 | { |
||
| 871 | "name": "stderr", |
||
| 872 | "output_type": "stream", |
||
| 873 | "text": [ |
||
| 874 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 875 | ] |
||
| 876 | }, |
||
| 877 | { |
||
| 878 | "name": "stdout", |
||
| 879 | "output_type": "stream", |
||
| 880 | "text": [ |
||
| 881 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 882 | ] |
||
| 883 | }, |
||
| 884 | { |
||
| 885 | "name": "stderr", |
||
| 886 | "output_type": "stream", |
||
| 887 | "text": [ |
||
| 888 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 889 | ] |
||
| 890 | }, |
||
| 891 | { |
||
| 892 | "name": "stdout", |
||
| 893 | "output_type": "stream", |
||
| 894 | "text": [ |
||
| 895 | "500/500 [==============================] - 82s 165ms/step - loss: 0.6931 - accuracy: 0.5096 - val_loss: 0.6933 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 896 | "Epoch 8/20\n", |
||
| 897 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.5032WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 898 | ] |
||
| 899 | }, |
||
| 900 | { |
||
| 901 | "name": "stderr", |
||
| 902 | "output_type": "stream", |
||
| 903 | "text": [ |
||
| 904 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 905 | ] |
||
| 906 | }, |
||
| 907 | { |
||
| 908 | "name": "stdout", |
||
| 909 | "output_type": "stream", |
||
| 910 | "text": [ |
||
| 911 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 912 | ] |
||
| 913 | }, |
||
| 914 | { |
||
| 915 | "name": "stderr", |
||
| 916 | "output_type": "stream", |
||
| 917 | "text": [ |
||
| 918 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 919 | ] |
||
| 920 | }, |
||
| 921 | { |
||
| 922 | "name": "stdout", |
||
| 923 | "output_type": "stream", |
||
| 924 | "text": [ |
||
| 925 | "500/500 [==============================] - 81s 162ms/step - loss: 0.6932 - accuracy: 0.5032 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 926 | "Epoch 9/20\n", |
||
| 927 | "500/500 [==============================] - ETA: 0s - loss: 0.6933 - accuracy: 0.5006WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 928 | ] |
||
| 929 | }, |
||
| 930 | { |
||
| 931 | "name": "stderr", |
||
| 932 | "output_type": "stream", |
||
| 933 | "text": [ |
||
| 934 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 935 | ] |
||
| 936 | }, |
||
| 937 | { |
||
| 938 | "name": "stdout", |
||
| 939 | "output_type": "stream", |
||
| 940 | "text": [ |
||
| 941 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 942 | ] |
||
| 943 | }, |
||
| 944 | { |
||
| 945 | "name": "stderr", |
||
| 946 | "output_type": "stream", |
||
| 947 | "text": [ |
||
| 948 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 949 | ] |
||
| 950 | }, |
||
| 951 | { |
||
| 952 | "name": "stdout", |
||
| 953 | "output_type": "stream", |
||
| 954 | "text": [ |
||
| 955 | "500/500 [==============================] - 81s 161ms/step - loss: 0.6933 - accuracy: 0.5006 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 956 | "Epoch 10/20\n", |
||
| 957 | "500/500 [==============================] - ETA: 0s - loss: 0.6930 - accuracy: 0.5150WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 958 | ] |
||
| 959 | }, |
||
| 960 | { |
||
| 961 | "name": "stderr", |
||
| 962 | "output_type": "stream", |
||
| 963 | "text": [ |
||
| 964 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 965 | ] |
||
| 966 | }, |
||
| 967 | { |
||
| 968 | "name": "stdout", |
||
| 969 | "output_type": "stream", |
||
| 970 | "text": [ |
||
| 971 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 972 | ] |
||
| 973 | }, |
||
| 974 | { |
||
| 975 | "name": "stderr", |
||
| 976 | "output_type": "stream", |
||
| 977 | "text": [ |
||
| 978 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 979 | ] |
||
| 980 | }, |
||
| 981 | { |
||
| 982 | "name": "stdout", |
||
| 983 | "output_type": "stream", |
||
| 984 | "text": [ |
||
| 985 | "500/500 [==============================] - 81s 163ms/step - loss: 0.6930 - accuracy: 0.5150 - val_loss: 0.6933 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 986 | "Epoch 11/20\n", |
||
| 987 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.5024WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 988 | ] |
||
| 989 | }, |
||
| 990 | { |
||
| 991 | "name": "stderr", |
||
| 992 | "output_type": "stream", |
||
| 993 | "text": [ |
||
| 994 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 995 | ] |
||
| 996 | }, |
||
| 997 | { |
||
| 998 | "name": "stdout", |
||
| 999 | "output_type": "stream", |
||
| 1000 | "text": [ |
||
| 1001 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1002 | ] |
||
| 1003 | }, |
||
| 1004 | { |
||
| 1005 | "name": "stderr", |
||
| 1006 | "output_type": "stream", |
||
| 1007 | "text": [ |
||
| 1008 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1009 | ] |
||
| 1010 | }, |
||
| 1011 | { |
||
| 1012 | "name": "stdout", |
||
| 1013 | "output_type": "stream", |
||
| 1014 | "text": [ |
||
| 1015 | "500/500 [==============================] - 84s 168ms/step - loss: 0.6932 - accuracy: 0.5024 - val_loss: 0.6933 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1016 | "Epoch 12/20\n", |
||
| 1017 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.5018WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1018 | ] |
||
| 1019 | }, |
||
| 1020 | { |
||
| 1021 | "name": "stderr", |
||
| 1022 | "output_type": "stream", |
||
| 1023 | "text": [ |
||
| 1024 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1025 | ] |
||
| 1026 | }, |
||
| 1027 | { |
||
| 1028 | "name": "stdout", |
||
| 1029 | "output_type": "stream", |
||
| 1030 | "text": [ |
||
| 1031 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1032 | ] |
||
| 1033 | }, |
||
| 1034 | { |
||
| 1035 | "name": "stderr", |
||
| 1036 | "output_type": "stream", |
||
| 1037 | "text": [ |
||
| 1038 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1039 | ] |
||
| 1040 | }, |
||
| 1041 | { |
||
| 1042 | "name": "stdout", |
||
| 1043 | "output_type": "stream", |
||
| 1044 | "text": [ |
||
| 1045 | "500/500 [==============================] - 88s 176ms/step - loss: 0.6932 - accuracy: 0.5018 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1046 | "Epoch 13/20\n", |
||
| 1047 | "500/500 [==============================] - ETA: 0s - loss: 0.6931 - accuracy: 0.5094WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1048 | ] |
||
| 1049 | }, |
||
| 1050 | { |
||
| 1051 | "name": "stderr", |
||
| 1052 | "output_type": "stream", |
||
| 1053 | "text": [ |
||
| 1054 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1055 | ] |
||
| 1056 | }, |
||
| 1057 | { |
||
| 1058 | "name": "stdout", |
||
| 1059 | "output_type": "stream", |
||
| 1060 | "text": [ |
||
| 1061 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1062 | ] |
||
| 1063 | }, |
||
| 1064 | { |
||
| 1065 | "name": "stderr", |
||
| 1066 | "output_type": "stream", |
||
| 1067 | "text": [ |
||
| 1068 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1069 | ] |
||
| 1070 | }, |
||
| 1071 | { |
||
| 1072 | "name": "stdout", |
||
| 1073 | "output_type": "stream", |
||
| 1074 | "text": [ |
||
| 1075 | "500/500 [==============================] - 88s 176ms/step - loss: 0.6931 - accuracy: 0.5094 - val_loss: 0.6933 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1076 | "Epoch 14/20\n", |
||
| 1077 | "500/500 [==============================] - ETA: 0s - loss: 0.6933 - accuracy: 0.4952WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1078 | ] |
||
| 1079 | }, |
||
| 1080 | { |
||
| 1081 | "name": "stderr", |
||
| 1082 | "output_type": "stream", |
||
| 1083 | "text": [ |
||
| 1084 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1085 | ] |
||
| 1086 | }, |
||
| 1087 | { |
||
| 1088 | "name": "stdout", |
||
| 1089 | "output_type": "stream", |
||
| 1090 | "text": [ |
||
| 1091 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1092 | ] |
||
| 1093 | }, |
||
| 1094 | { |
||
| 1095 | "name": "stderr", |
||
| 1096 | "output_type": "stream", |
||
| 1097 | "text": [ |
||
| 1098 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1099 | ] |
||
| 1100 | }, |
||
| 1101 | { |
||
| 1102 | "name": "stdout", |
||
| 1103 | "output_type": "stream", |
||
| 1104 | "text": [ |
||
| 1105 | "500/500 [==============================] - 87s 173ms/step - loss: 0.6933 - accuracy: 0.4952 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1106 | "Epoch 15/20\n", |
||
| 1107 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.5010WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1108 | ] |
||
| 1109 | }, |
||
| 1110 | { |
||
| 1111 | "name": "stderr", |
||
| 1112 | "output_type": "stream", |
||
| 1113 | "text": [ |
||
| 1114 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1115 | ] |
||
| 1116 | }, |
||
| 1117 | { |
||
| 1118 | "name": "stdout", |
||
| 1119 | "output_type": "stream", |
||
| 1120 | "text": [ |
||
| 1121 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1122 | ] |
||
| 1123 | }, |
||
| 1124 | { |
||
| 1125 | "name": "stderr", |
||
| 1126 | "output_type": "stream", |
||
| 1127 | "text": [ |
||
| 1128 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1129 | ] |
||
| 1130 | }, |
||
| 1131 | { |
||
| 1132 | "name": "stdout", |
||
| 1133 | "output_type": "stream", |
||
| 1134 | "text": [ |
||
| 1135 | "500/500 [==============================] - 84s 167ms/step - loss: 0.6932 - accuracy: 0.5010 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1136 | "Epoch 16/20\n", |
||
| 1137 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.4996WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1138 | ] |
||
| 1139 | }, |
||
| 1140 | { |
||
| 1141 | "name": "stderr", |
||
| 1142 | "output_type": "stream", |
||
| 1143 | "text": [ |
||
| 1144 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1145 | ] |
||
| 1146 | }, |
||
| 1147 | { |
||
| 1148 | "name": "stdout", |
||
| 1149 | "output_type": "stream", |
||
| 1150 | "text": [ |
||
| 1151 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1152 | ] |
||
| 1153 | }, |
||
| 1154 | { |
||
| 1155 | "name": "stderr", |
||
| 1156 | "output_type": "stream", |
||
| 1157 | "text": [ |
||
| 1158 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1159 | ] |
||
| 1160 | }, |
||
| 1161 | { |
||
| 1162 | "name": "stdout", |
||
| 1163 | "output_type": "stream", |
||
| 1164 | "text": [ |
||
| 1165 | "500/500 [==============================] - 84s 167ms/step - loss: 0.6932 - accuracy: 0.4996 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1166 | "Epoch 17/20\n", |
||
| 1167 | "500/500 [==============================] - ETA: 0s - loss: 0.6932 - accuracy: 0.5046WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1168 | ] |
||
| 1169 | }, |
||
| 1170 | { |
||
| 1171 | "name": "stderr", |
||
| 1172 | "output_type": "stream", |
||
| 1173 | "text": [ |
||
| 1174 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1175 | ] |
||
| 1176 | }, |
||
| 1177 | { |
||
| 1178 | "name": "stdout", |
||
| 1179 | "output_type": "stream", |
||
| 1180 | "text": [ |
||
| 1181 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1182 | ] |
||
| 1183 | }, |
||
| 1184 | { |
||
| 1185 | "name": "stderr", |
||
| 1186 | "output_type": "stream", |
||
| 1187 | "text": [ |
||
| 1188 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1189 | ] |
||
| 1190 | }, |
||
| 1191 | { |
||
| 1192 | "name": "stdout", |
||
| 1193 | "output_type": "stream", |
||
| 1194 | "text": [ |
||
| 1195 | "500/500 [==============================] - 82s 164ms/step - loss: 0.6932 - accuracy: 0.5046 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1196 | "Epoch 18/20\n", |
||
| 1197 | "500/500 [==============================] - ETA: 0s - loss: 0.6931 - accuracy: 0.5070WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1198 | ] |
||
| 1199 | }, |
||
| 1200 | { |
||
| 1201 | "name": "stderr", |
||
| 1202 | "output_type": "stream", |
||
| 1203 | "text": [ |
||
| 1204 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1205 | ] |
||
| 1206 | }, |
||
| 1207 | { |
||
| 1208 | "name": "stdout", |
||
| 1209 | "output_type": "stream", |
||
| 1210 | "text": [ |
||
| 1211 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1212 | ] |
||
| 1213 | }, |
||
| 1214 | { |
||
| 1215 | "name": "stderr", |
||
| 1216 | "output_type": "stream", |
||
| 1217 | "text": [ |
||
| 1218 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1219 | ] |
||
| 1220 | }, |
||
| 1221 | { |
||
| 1222 | "name": "stdout", |
||
| 1223 | "output_type": "stream", |
||
| 1224 | "text": [ |
||
| 1225 | "500/500 [==============================] - 81s 162ms/step - loss: 0.6931 - accuracy: 0.5070 - val_loss: 0.6932 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1226 | "Epoch 19/20\n", |
||
| 1227 | "500/500 [==============================] - ETA: 0s - loss: 0.6930 - accuracy: 0.5106WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1228 | ] |
||
| 1229 | }, |
||
| 1230 | { |
||
| 1231 | "name": "stderr", |
||
| 1232 | "output_type": "stream", |
||
| 1233 | "text": [ |
||
| 1234 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1235 | ] |
||
| 1236 | }, |
||
| 1237 | { |
||
| 1238 | "name": "stdout", |
||
| 1239 | "output_type": "stream", |
||
| 1240 | "text": [ |
||
| 1241 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1242 | ] |
||
| 1243 | }, |
||
| 1244 | { |
||
| 1245 | "name": "stderr", |
||
| 1246 | "output_type": "stream", |
||
| 1247 | "text": [ |
||
| 1248 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1249 | ] |
||
| 1250 | }, |
||
| 1251 | { |
||
| 1252 | "name": "stdout", |
||
| 1253 | "output_type": "stream", |
||
| 1254 | "text": [ |
||
| 1255 | "500/500 [==============================] - 81s 162ms/step - loss: 0.6930 - accuracy: 0.5106 - val_loss: 0.6933 - val_accuracy: 0.5000 - lr: 0.0010\n", |
||
| 1256 | "Epoch 20/20\n", |
||
| 1257 | "500/500 [==============================] - ETA: 0s - loss: 0.6929 - accuracy: 0.5126WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1258 | ] |
||
| 1259 | }, |
||
| 1260 | { |
||
| 1261 | "name": "stderr", |
||
| 1262 | "output_type": "stream", |
||
| 1263 | "text": [ |
||
| 1264 | "WARNING:tensorflow:Can save best model only with val_acc available, skipping.\n" |
||
| 1265 | ] |
||
| 1266 | }, |
||
| 1267 | { |
||
| 1268 | "name": "stdout", |
||
| 1269 | "output_type": "stream", |
||
| 1270 | "text": [ |
||
| 1271 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1272 | ] |
||
| 1273 | }, |
||
| 1274 | { |
||
| 1275 | "name": "stderr", |
||
| 1276 | "output_type": "stream", |
||
| 1277 | "text": [ |
||
| 1278 | "WARNING:tensorflow:Learning rate reduction is conditioned on metric `val_acc` which is not available. Available metrics are: loss,accuracy,val_loss,val_accuracy,lr\n" |
||
| 1279 | ] |
||
| 1280 | }, |
||
| 1281 | { |
||
| 1282 | "name": "stdout", |
||
| 1283 | "output_type": "stream", |
||
| 1284 | "text": [ |
||
| 1285 | "500/500 [==============================] - 82s 165ms/step - loss: 0.6929 - accuracy: 0.5126 - val_loss: 0.6934 - val_accuracy: 0.5000 - lr: 0.0010\n" |
||
| 1286 | ] |
||
| 1287 | } |
||
| 1288 | ], |
||
| 1289 | "source": [ |
||
| 1290 | "filepath = \"model.h5\"\n", |
||
| 1291 | "checkpoint = ModelCheckpoint(filepath, monitor='val_acc', verbose=1, \n", |
||
| 1292 | " save_best_only=True, mode='max')\n", |
||
| 1293 | "\n", |
||
| 1294 | "reduce_lr = ReduceLROnPlateau(monitor='val_acc', factor=0.5, patience=2, \n", |
||
| 1295 | " verbose=1, mode='max', min_lr=0.00001)\n", |
||
| 1296 | " \n", |
||
| 1297 | " \n", |
||
| 1298 | "callbacks_list = [checkpoint, reduce_lr]\n", |
||
| 1299 | "\n", |
||
| 1300 | "history = model.fit_generator(train_gen, steps_per_epoch=train_steps, \n", |
||
| 1301 | " validation_data=val_gen,\n", |
||
| 1302 | " validation_steps=val_steps,\n", |
||
| 1303 | " epochs=20, verbose=1,\n", |
||
| 1304 | " callbacks=callbacks_list)" |
||
| 1305 | ] |
||
| 1306 | }, |
||
| 1307 | { |
||
| 1308 | "cell_type": "code", |
||
| 1309 | "execution_count": 31, |
||
| 1310 | "metadata": {}, |
||
| 1311 | "outputs": [ |
||
| 1312 | { |
||
| 1313 | "data": { |
||
| 1314 | "text/plain": [ |
||
| 1315 | "['loss', 'accuracy']" |
||
| 1316 | ] |
||
| 1317 | }, |
||
| 1318 | "execution_count": 31, |
||
| 1319 | "metadata": {}, |
||
| 1320 | "output_type": "execute_result" |
||
| 1321 | } |
||
| 1322 | ], |
||
| 1323 | "source": [ |
||
| 1324 | "# get the metric names so we can use evaulate_generator\n", |
||
| 1325 | "model.metrics_names" |
||
| 1326 | ] |
||
| 1327 | }, |
||
| 1328 | { |
||
| 1329 | "cell_type": "code", |
||
| 1330 | "execution_count": 32, |
||
| 1331 | "metadata": {}, |
||
| 1332 | "outputs": [ |
||
| 1333 | { |
||
| 1334 | "name": "stderr", |
||
| 1335 | "output_type": "stream", |
||
| 1336 | "text": [ |
||
| 1337 | "/var/folders/3k/vhmk7kx14ybb600l38fbqg780000gn/T/ipykernel_68365/2341706544.py:6: UserWarning: `Model.evaluate_generator` is deprecated and will be removed in a future version. Please use `Model.evaluate`, which supports generators.\n", |
||
| 1338 | " model.evaluate_generator(test_gen,\n" |
||
| 1339 | ] |
||
| 1340 | }, |
||
| 1341 | { |
||
| 1342 | "name": "stdout", |
||
| 1343 | "output_type": "stream", |
||
| 1344 | "text": [ |
||
| 1345 | "val_loss: 0.6933144330978394\n", |
||
| 1346 | "val_acc: 0.5\n" |
||
| 1347 | ] |
||
| 1348 | } |
||
| 1349 | ], |
||
| 1350 | "source": [ |
||
| 1351 | "# Here the best epoch will be used.\n", |
||
| 1352 | "\n", |
||
| 1353 | "\n", |
||
| 1354 | "\n", |
||
| 1355 | "val_loss, val_acc = \\\n", |
||
| 1356 | "model.evaluate_generator(test_gen, \n", |
||
| 1357 | " steps=len(df_val))\n", |
||
| 1358 | "\n", |
||
| 1359 | "print('val_loss:', val_loss)\n", |
||
| 1360 | "print('val_acc:', val_acc)" |
||
| 1361 | ] |
||
| 1362 | }, |
||
| 1363 | { |
||
| 1364 | "cell_type": "code", |
||
| 1365 | "execution_count": 33, |
||
| 1366 | "metadata": {}, |
||
| 1367 | "outputs": [ |
||
| 1368 | { |
||
| 1369 | "data": { |
||
| 1370 | "text/plain": [ |
||
| 1371 | "<Figure size 640x480 with 0 Axes>" |
||
| 1372 | ] |
||
| 1373 | }, |
||
| 1374 | "execution_count": 33, |
||
| 1375 | "metadata": {}, |
||
| 1376 | "output_type": "execute_result" |
||
| 1377 | }, |
||
| 1378 | { |
||
| 1379 | "data": { |
||
| 1380 | "image/png": "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\n", |
||
| 1381 | "text/plain": [ |
||
| 1382 | "<Figure size 640x480 with 1 Axes>" |
||
| 1383 | ] |
||
| 1384 | }, |
||
| 1385 | "metadata": {}, |
||
| 1386 | "output_type": "display_data" |
||
| 1387 | }, |
||
| 1388 | { |
||
| 1389 | "data": { |
||
| 1390 | "image/png": "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\n", |
||
| 1391 | "text/plain": [ |
||
| 1392 | "<Figure size 640x480 with 1 Axes>" |
||
| 1393 | ] |
||
| 1394 | }, |
||
| 1395 | "metadata": {}, |
||
| 1396 | "output_type": "display_data" |
||
| 1397 | }, |
||
| 1398 | { |
||
| 1399 | "data": { |
||
| 1400 | "text/plain": [ |
||
| 1401 | "<Figure size 640x480 with 0 Axes>" |
||
| 1402 | ] |
||
| 1403 | }, |
||
| 1404 | "metadata": {}, |
||
| 1405 | "output_type": "display_data" |
||
| 1406 | } |
||
| 1407 | ], |
||
| 1408 | "source": [ |
||
| 1409 | "# display the loss and accuracy curves\n", |
||
| 1410 | "\n", |
||
| 1411 | "import matplotlib.pyplot as plt\n", |
||
| 1412 | "\n", |
||
| 1413 | "acc = history.history['accuracy']\n", |
||
| 1414 | "val_acc = history.history['val_accuracy']\n", |
||
| 1415 | "loss = history.history['loss']\n", |
||
| 1416 | "val_loss = history.history['val_loss']\n", |
||
| 1417 | "\n", |
||
| 1418 | "epochs = range(1, len(acc) + 1)\n", |
||
| 1419 | "\n", |
||
| 1420 | "plt.plot(epochs, loss, 'bo', label='Training loss')\n", |
||
| 1421 | "plt.plot(epochs, val_loss, 'b', label='Validation loss')\n", |
||
| 1422 | "plt.title('Training and validation loss')\n", |
||
| 1423 | "plt.legend()\n", |
||
| 1424 | "plt.figure()\n", |
||
| 1425 | "\n", |
||
| 1426 | "plt.plot(epochs, acc, 'bo', label='Training acc')\n", |
||
| 1427 | "plt.plot(epochs, val_acc, 'b', label='Validation acc')\n", |
||
| 1428 | "plt.title('Training and validation accuracy')\n", |
||
| 1429 | "plt.legend()\n", |
||
| 1430 | "plt.figure()" |
||
| 1431 | ] |
||
| 1432 | }, |
||
| 1433 | { |
||
| 1434 | "cell_type": "markdown", |
||
| 1435 | "metadata": {}, |
||
| 1436 | "source": [ |
||
| 1437 | "# 4. Validation and Analysis \n", |
||
| 1438 | "\n", |
||
| 1439 | "* ### Metrics\n", |
||
| 1440 | "* ### Prediction and Activation Visualizations\n", |
||
| 1441 | "* ### ROC and AUC" |
||
| 1442 | ] |
||
| 1443 | }, |
||
| 1444 | { |
||
| 1445 | "cell_type": "code", |
||
| 1446 | "execution_count": 34, |
||
| 1447 | "metadata": {}, |
||
| 1448 | "outputs": [ |
||
| 1449 | { |
||
| 1450 | "name": "stderr", |
||
| 1451 | "output_type": "stream", |
||
| 1452 | "text": [ |
||
| 1453 | "/var/folders/3k/vhmk7kx14ybb600l38fbqg780000gn/T/ipykernel_68365/2435426349.py:2: UserWarning: `Model.predict_generator` is deprecated and will be removed in a future version. Please use `Model.predict`, which supports generators.\n", |
||
| 1454 | " predictions = model.predict_generator(test_gen, steps=len(df_val), verbose=1)\n" |
||
| 1455 | ] |
||
| 1456 | }, |
||
| 1457 | { |
||
| 1458 | "name": "stdout", |
||
| 1459 | "output_type": "stream", |
||
| 1460 | "text": [ |
||
| 1461 | "16000/16000 [==============================] - 73s 5ms/step\n" |
||
| 1462 | ] |
||
| 1463 | } |
||
| 1464 | ], |
||
| 1465 | "source": [ |
||
| 1466 | "# make a prediction\n", |
||
| 1467 | "predictions = model.predict_generator(test_gen, steps=len(df_val), verbose=1)" |
||
| 1468 | ] |
||
| 1469 | }, |
||
| 1470 | { |
||
| 1471 | "cell_type": "code", |
||
| 1472 | "execution_count": 35, |
||
| 1473 | "metadata": {}, |
||
| 1474 | "outputs": [ |
||
| 1475 | { |
||
| 1476 | "data": { |
||
| 1477 | "text/plain": [ |
||
| 1478 | "(16000, 2)" |
||
| 1479 | ] |
||
| 1480 | }, |
||
| 1481 | "execution_count": 35, |
||
| 1482 | "metadata": {}, |
||
| 1483 | "output_type": "execute_result" |
||
| 1484 | } |
||
| 1485 | ], |
||
| 1486 | "source": [ |
||
| 1487 | "predictions.shape" |
||
| 1488 | ] |
||
| 1489 | }, |
||
| 1490 | { |
||
| 1491 | "cell_type": "code", |
||
| 1492 | "execution_count": 36, |
||
| 1493 | "metadata": {}, |
||
| 1494 | "outputs": [ |
||
| 1495 | { |
||
| 1496 | "data": { |
||
| 1497 | "text/plain": [ |
||
| 1498 | "{'a_no_tumor_tissue': 0, 'b_has_tumor_tissue': 1}" |
||
| 1499 | ] |
||
| 1500 | }, |
||
| 1501 | "execution_count": 36, |
||
| 1502 | "metadata": {}, |
||
| 1503 | "output_type": "execute_result" |
||
| 1504 | } |
||
| 1505 | ], |
||
| 1506 | "source": [ |
||
| 1507 | "# This is how to check what index keras has internally assigned to each class. \n", |
||
| 1508 | "test_gen.class_indices" |
||
| 1509 | ] |
||
| 1510 | }, |
||
| 1511 | { |
||
| 1512 | "cell_type": "code", |
||
| 1513 | "execution_count": 37, |
||
| 1514 | "metadata": {}, |
||
| 1515 | "outputs": [ |
||
| 1516 | { |
||
| 1517 | "data": { |
||
| 1518 | "text/html": [ |
||
| 1519 | "<div>\n", |
||
| 1520 | "<style scoped>\n", |
||
| 1521 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 1522 | " vertical-align: middle;\n", |
||
| 1523 | " }\n", |
||
| 1524 | "\n", |
||
| 1525 | " .dataframe tbody tr th {\n", |
||
| 1526 | " vertical-align: top;\n", |
||
| 1527 | " }\n", |
||
| 1528 | "\n", |
||
| 1529 | " .dataframe thead th {\n", |
||
| 1530 | " text-align: right;\n", |
||
| 1531 | " }\n", |
||
| 1532 | "</style>\n", |
||
| 1533 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 1534 | " <thead>\n", |
||
| 1535 | " <tr style=\"text-align: right;\">\n", |
||
| 1536 | " <th></th>\n", |
||
| 1537 | " <th>no_tumor_tissue</th>\n", |
||
| 1538 | " <th>has_tumor_tissue</th>\n", |
||
| 1539 | " </tr>\n", |
||
| 1540 | " </thead>\n", |
||
| 1541 | " <tbody>\n", |
||
| 1542 | " <tr>\n", |
||
| 1543 | " <th>0</th>\n", |
||
| 1544 | " <td>0.477844</td>\n", |
||
| 1545 | " <td>0.522156</td>\n", |
||
| 1546 | " </tr>\n", |
||
| 1547 | " <tr>\n", |
||
| 1548 | " <th>1</th>\n", |
||
| 1549 | " <td>0.477844</td>\n", |
||
| 1550 | " <td>0.522156</td>\n", |
||
| 1551 | " </tr>\n", |
||
| 1552 | " <tr>\n", |
||
| 1553 | " <th>2</th>\n", |
||
| 1554 | " <td>0.477844</td>\n", |
||
| 1555 | " <td>0.522156</td>\n", |
||
| 1556 | " </tr>\n", |
||
| 1557 | " <tr>\n", |
||
| 1558 | " <th>3</th>\n", |
||
| 1559 | " <td>0.477844</td>\n", |
||
| 1560 | " <td>0.522156</td>\n", |
||
| 1561 | " </tr>\n", |
||
| 1562 | " <tr>\n", |
||
| 1563 | " <th>4</th>\n", |
||
| 1564 | " <td>0.477844</td>\n", |
||
| 1565 | " <td>0.522156</td>\n", |
||
| 1566 | " </tr>\n", |
||
| 1567 | " </tbody>\n", |
||
| 1568 | "</table>\n", |
||
| 1569 | "</div>" |
||
| 1570 | ], |
||
| 1571 | "text/plain": [ |
||
| 1572 | " no_tumor_tissue has_tumor_tissue\n", |
||
| 1573 | "0 0.477844 0.522156\n", |
||
| 1574 | "1 0.477844 0.522156\n", |
||
| 1575 | "2 0.477844 0.522156\n", |
||
| 1576 | "3 0.477844 0.522156\n", |
||
| 1577 | "4 0.477844 0.522156" |
||
| 1578 | ] |
||
| 1579 | }, |
||
| 1580 | "execution_count": 37, |
||
| 1581 | "metadata": {}, |
||
| 1582 | "output_type": "execute_result" |
||
| 1583 | } |
||
| 1584 | ], |
||
| 1585 | "source": [ |
||
| 1586 | "# Put the predictions into a dataframe.\n", |
||
| 1587 | "# The columns need to be ordered to match the output of the previous cell\n", |
||
| 1588 | "\n", |
||
| 1589 | "df_preds = pd.DataFrame(predictions, columns=['no_tumor_tissue', 'has_tumor_tissue'])\n", |
||
| 1590 | "\n", |
||
| 1591 | "df_preds.head()\n" |
||
| 1592 | ] |
||
| 1593 | }, |
||
| 1594 | { |
||
| 1595 | "cell_type": "code", |
||
| 1596 | "execution_count": 38, |
||
| 1597 | "metadata": {}, |
||
| 1598 | "outputs": [], |
||
| 1599 | "source": [ |
||
| 1600 | "# Get the true labels\n", |
||
| 1601 | "y_true = test_gen.classes\n", |
||
| 1602 | "\n", |
||
| 1603 | "# Get the predicted labels as probabilities\n", |
||
| 1604 | "y_pred = df_preds['has_tumor_tissue']" |
||
| 1605 | ] |
||
| 1606 | }, |
||
| 1607 | { |
||
| 1608 | "cell_type": "code", |
||
| 1609 | "execution_count": 39, |
||
| 1610 | "metadata": {}, |
||
| 1611 | "outputs": [ |
||
| 1612 | { |
||
| 1613 | "data": { |
||
| 1614 | "text/plain": [ |
||
| 1615 | "0.5" |
||
| 1616 | ] |
||
| 1617 | }, |
||
| 1618 | "execution_count": 39, |
||
| 1619 | "metadata": {}, |
||
| 1620 | "output_type": "execute_result" |
||
| 1621 | } |
||
| 1622 | ], |
||
| 1623 | "source": [ |
||
| 1624 | "from sklearn.metrics import roc_auc_score\n", |
||
| 1625 | "\n", |
||
| 1626 | "roc_auc_score(y_true, y_pred)" |
||
| 1627 | ] |
||
| 1628 | }, |
||
| 1629 | { |
||
| 1630 | "cell_type": "code", |
||
| 1631 | "execution_count": 40, |
||
| 1632 | "metadata": {}, |
||
| 1633 | "outputs": [ |
||
| 1634 | { |
||
| 1635 | "data": { |
||
| 1636 | "text/plain": [ |
||
| 1637 | "(16000,)" |
||
| 1638 | ] |
||
| 1639 | }, |
||
| 1640 | "execution_count": 40, |
||
| 1641 | "metadata": {}, |
||
| 1642 | "output_type": "execute_result" |
||
| 1643 | } |
||
| 1644 | ], |
||
| 1645 | "source": [ |
||
| 1646 | "# Get the labels of the test images.\n", |
||
| 1647 | "\n", |
||
| 1648 | "test_labels = test_gen.classes\n", |
||
| 1649 | "test_labels.shape" |
||
| 1650 | ] |
||
| 1651 | }, |
||
| 1652 | { |
||
| 1653 | "cell_type": "code", |
||
| 1654 | "execution_count": 41, |
||
| 1655 | "metadata": {}, |
||
| 1656 | "outputs": [ |
||
| 1657 | { |
||
| 1658 | "data": { |
||
| 1659 | "text/plain": [ |
||
| 1660 | "{'a_no_tumor_tissue': 0, 'b_has_tumor_tissue': 1}" |
||
| 1661 | ] |
||
| 1662 | }, |
||
| 1663 | "execution_count": 41, |
||
| 1664 | "metadata": {}, |
||
| 1665 | "output_type": "execute_result" |
||
| 1666 | } |
||
| 1667 | ], |
||
| 1668 | "source": [ |
||
| 1669 | "# argmax returns the index of the max value in a row\n", |
||
| 1670 | "cm = confusion_matrix(test_labels, predictions.argmax(axis=1))\n", |
||
| 1671 | "# Print the label associated with each class\n", |
||
| 1672 | "test_gen.class_indices" |
||
| 1673 | ] |
||
| 1674 | }, |
||
| 1675 | { |
||
| 1676 | "cell_type": "code", |
||
| 1677 | "execution_count": 43, |
||
| 1678 | "metadata": {}, |
||
| 1679 | "outputs": [ |
||
| 1680 | { |
||
| 1681 | "name": "stdout", |
||
| 1682 | "output_type": "stream", |
||
| 1683 | "text": [ |
||
| 1684 | "Collecting sklearn\n", |
||
| 1685 | " Downloading sklearn-0.0.post7.tar.gz (3.6 kB)\n", |
||
| 1686 | " Preparing metadata (setup.py) ... \u001b[?25lerror\n", |
||
| 1687 | " \u001b[1;31merror\u001b[0m: \u001b[1msubprocess-exited-with-error\u001b[0m\n", |
||
| 1688 | " \n", |
||
| 1689 | " \u001b[31m×\u001b[0m \u001b[32mpython setup.py egg_info\u001b[0m did not run successfully.\n", |
||
| 1690 | " \u001b[31m│\u001b[0m exit code: \u001b[1;36m1\u001b[0m\n", |
||
| 1691 | " \u001b[31m╰─>\u001b[0m \u001b[31m[18 lines of output]\u001b[0m\n", |
||
| 1692 | " \u001b[31m \u001b[0m The 'sklearn' PyPI package is deprecated, use 'scikit-learn'\n", |
||
| 1693 | " \u001b[31m \u001b[0m rather than 'sklearn' for pip commands.\n", |
||
| 1694 | " \u001b[31m \u001b[0m \n", |
||
| 1695 | " \u001b[31m \u001b[0m Here is how to fix this error in the main use cases:\n", |
||
| 1696 | " \u001b[31m \u001b[0m - use 'pip install scikit-learn' rather than 'pip install sklearn'\n", |
||
| 1697 | " \u001b[31m \u001b[0m - replace 'sklearn' by 'scikit-learn' in your pip requirements files\n", |
||
| 1698 | " \u001b[31m \u001b[0m (requirements.txt, setup.py, setup.cfg, Pipfile, etc ...)\n", |
||
| 1699 | " \u001b[31m \u001b[0m - if the 'sklearn' package is used by one of your dependencies,\n", |
||
| 1700 | " \u001b[31m \u001b[0m it would be great if you take some time to track which package uses\n", |
||
| 1701 | " \u001b[31m \u001b[0m 'sklearn' instead of 'scikit-learn' and report it to their issue tracker\n", |
||
| 1702 | " \u001b[31m \u001b[0m - as a last resort, set the environment variable\n", |
||
| 1703 | " \u001b[31m \u001b[0m SKLEARN_ALLOW_DEPRECATED_SKLEARN_PACKAGE_INSTALL=True to avoid this error\n", |
||
| 1704 | " \u001b[31m \u001b[0m \n", |
||
| 1705 | " \u001b[31m \u001b[0m More information is available at\n", |
||
| 1706 | " \u001b[31m \u001b[0m https://github.com/scikit-learn/sklearn-pypi-package\n", |
||
| 1707 | " \u001b[31m \u001b[0m \n", |
||
| 1708 | " \u001b[31m \u001b[0m If the previous advice does not cover your use case, feel free to report it at\n", |
||
| 1709 | " \u001b[31m \u001b[0m https://github.com/scikit-learn/sklearn-pypi-package/issues/new\n", |
||
| 1710 | " \u001b[31m \u001b[0m \u001b[31m[end of output]\u001b[0m\n", |
||
| 1711 | " \n", |
||
| 1712 | " \u001b[1;35mnote\u001b[0m: This error originates from a subprocess, and is likely not a problem with pip.\n", |
||
| 1713 | "\u001b[1;31merror\u001b[0m: \u001b[1mmetadata-generation-failed\u001b[0m\n", |
||
| 1714 | "\n", |
||
| 1715 | "\u001b[31m×\u001b[0m Encountered error while generating package metadata.\n", |
||
| 1716 | "\u001b[31m╰─>\u001b[0m See above for output.\n", |
||
| 1717 | "\n", |
||
| 1718 | "\u001b[1;35mnote\u001b[0m: This is an issue with the package mentioned above, not pip.\n", |
||
| 1719 | "\u001b[1;36mhint\u001b[0m: See above for details.\n", |
||
| 1720 | "\u001b[?25h" |
||
| 1721 | ] |
||
| 1722 | } |
||
| 1723 | ], |
||
| 1724 | "source": [ |
||
| 1725 | "!pip install sklearn\n" |
||
| 1726 | ] |
||
| 1727 | }, |
||
| 1728 | { |
||
| 1729 | "cell_type": "code", |
||
| 1730 | "execution_count": 45, |
||
| 1731 | "metadata": {}, |
||
| 1732 | "outputs": [], |
||
| 1733 | "source": [ |
||
| 1734 | "# from sklearn.metrics import plot_confusion_matrix\n", |
||
| 1735 | "from sklearn.metrics import ConfusionMatrixDisplay" |
||
| 1736 | ] |
||
| 1737 | }, |
||
| 1738 | { |
||
| 1739 | "cell_type": "code", |
||
| 1740 | "execution_count": 46, |
||
| 1741 | "metadata": {}, |
||
| 1742 | "outputs": [ |
||
| 1743 | { |
||
| 1744 | "data": { |
||
| 1745 | "text/plain": [ |
||
| 1746 | "['test_images']" |
||
| 1747 | ] |
||
| 1748 | }, |
||
| 1749 | "execution_count": 46, |
||
| 1750 | "metadata": {}, |
||
| 1751 | "output_type": "execute_result" |
||
| 1752 | } |
||
| 1753 | ], |
||
| 1754 | "source": [ |
||
| 1755 | "# Delete base_dir and it's sub folders to free up disk space.\n", |
||
| 1756 | "\n", |
||
| 1757 | "shutil.rmtree('base_dir')\n", |
||
| 1758 | "#[CREATE A TEST FOLDER DIRECTORY STRUCTURE]\n", |
||
| 1759 | "\n", |
||
| 1760 | "# We will be feeding test images from a folder into predict_generator().\n", |
||
| 1761 | "# Keras requires that the path should point to a folder containing images and not\n", |
||
| 1762 | "# to the images themselves. That is why we are creating a folder (test_images) \n", |
||
| 1763 | "# inside another folder (test_dir).\n", |
||
| 1764 | "\n", |
||
| 1765 | "# test_dir\n", |
||
| 1766 | " # test_images\n", |
||
| 1767 | "\n", |
||
| 1768 | "# create test_dir\n", |
||
| 1769 | "test_dir = 'test_dir'\n", |
||
| 1770 | "os.mkdir(test_dir)\n", |
||
| 1771 | " \n", |
||
| 1772 | "# create test_images inside test_dir\n", |
||
| 1773 | "test_images = os.path.join(test_dir, 'test_images')\n", |
||
| 1774 | "os.mkdir(test_images)\n", |
||
| 1775 | "# check that the directory we created exists\n", |
||
| 1776 | "os.listdir('test_dir')" |
||
| 1777 | ] |
||
| 1778 | }, |
||
| 1779 | { |
||
| 1780 | "cell_type": "code", |
||
| 1781 | "execution_count": 49, |
||
| 1782 | "metadata": {}, |
||
| 1783 | "outputs": [ |
||
| 1784 | { |
||
| 1785 | "data": { |
||
| 1786 | "text/plain": [ |
||
| 1787 | "57458" |
||
| 1788 | ] |
||
| 1789 | }, |
||
| 1790 | "execution_count": 49, |
||
| 1791 | "metadata": {}, |
||
| 1792 | "output_type": "execute_result" |
||
| 1793 | } |
||
| 1794 | ], |
||
| 1795 | "source": [ |
||
| 1796 | "# Transfer the test images into image_dir\n", |
||
| 1797 | "\n", |
||
| 1798 | "test_list = os.listdir('./histopathologic-cancer-detection/test')\n", |
||
| 1799 | "\n", |
||
| 1800 | "for image in test_list:\n", |
||
| 1801 | " \n", |
||
| 1802 | " fname = image\n", |
||
| 1803 | " \n", |
||
| 1804 | " # source path to image\n", |
||
| 1805 | " src = os.path.join('./histopathologic-cancer-detection/test', fname)\n", |
||
| 1806 | " # destination path to image\n", |
||
| 1807 | " dst = os.path.join(test_images, fname)\n", |
||
| 1808 | " # copy the image from the source to the destination\n", |
||
| 1809 | " shutil.copyfile(src, dst)\n", |
||
| 1810 | "# check that the images are now in the test_images\n", |
||
| 1811 | "# Should now be 57458 images in the test_images folder\n", |
||
| 1812 | "\n", |
||
| 1813 | "len(os.listdir('test_dir/test_images'))" |
||
| 1814 | ] |
||
| 1815 | }, |
||
| 1816 | { |
||
| 1817 | "cell_type": "code", |
||
| 1818 | "execution_count": 50, |
||
| 1819 | "metadata": {}, |
||
| 1820 | "outputs": [ |
||
| 1821 | { |
||
| 1822 | "name": "stdout", |
||
| 1823 | "output_type": "stream", |
||
| 1824 | "text": [ |
||
| 1825 | "Found 57458 images belonging to 1 classes.\n" |
||
| 1826 | ] |
||
| 1827 | } |
||
| 1828 | ], |
||
| 1829 | "source": [ |
||
| 1830 | "test_path ='test_dir'\n", |
||
| 1831 | "\n", |
||
| 1832 | "\n", |
||
| 1833 | "# Here we change the path to point to the test_images folder.\n", |
||
| 1834 | "\n", |
||
| 1835 | "test_gen = datagen.flow_from_directory(test_path,\n", |
||
| 1836 | " target_size=(IMAGE_SIZE,IMAGE_SIZE),\n", |
||
| 1837 | " batch_size=1,\n", |
||
| 1838 | " class_mode='categorical',\n", |
||
| 1839 | " shuffle=False)" |
||
| 1840 | ] |
||
| 1841 | }, |
||
| 1842 | { |
||
| 1843 | "cell_type": "code", |
||
| 1844 | "execution_count": 51, |
||
| 1845 | "metadata": {}, |
||
| 1846 | "outputs": [ |
||
| 1847 | { |
||
| 1848 | "name": "stdout", |
||
| 1849 | "output_type": "stream", |
||
| 1850 | "text": [ |
||
| 1851 | " 23/57458 [..............................] - ETA: 4:34" |
||
| 1852 | ] |
||
| 1853 | }, |
||
| 1854 | { |
||
| 1855 | "name": "stderr", |
||
| 1856 | "output_type": "stream", |
||
| 1857 | "text": [ |
||
| 1858 | "/var/folders/3k/vhmk7kx14ybb600l38fbqg780000gn/T/ipykernel_68365/3697970497.py:5: UserWarning: `Model.predict_generator` is deprecated and will be removed in a future version. Please use `Model.predict`, which supports generators.\n", |
||
| 1859 | " predictions = model.predict_generator(test_gen, steps=num_test_images, verbose=1)\n" |
||
| 1860 | ] |
||
| 1861 | }, |
||
| 1862 | { |
||
| 1863 | "name": "stdout", |
||
| 1864 | "output_type": "stream", |
||
| 1865 | "text": [ |
||
| 1866 | "57458/57458 [==============================] - 244s 4ms/step\n" |
||
| 1867 | ] |
||
| 1868 | } |
||
| 1869 | ], |
||
| 1870 | "source": [ |
||
| 1871 | "num_test_images = 57458\n", |
||
| 1872 | "\n", |
||
| 1873 | "\n", |
||
| 1874 | "\n", |
||
| 1875 | "predictions = model.predict_generator(test_gen, steps=num_test_images, verbose=1)" |
||
| 1876 | ] |
||
| 1877 | }, |
||
| 1878 | { |
||
| 1879 | "cell_type": "code", |
||
| 1880 | "execution_count": 52, |
||
| 1881 | "metadata": {}, |
||
| 1882 | "outputs": [ |
||
| 1883 | { |
||
| 1884 | "data": { |
||
| 1885 | "text/plain": [ |
||
| 1886 | "57458" |
||
| 1887 | ] |
||
| 1888 | }, |
||
| 1889 | "execution_count": 52, |
||
| 1890 | "metadata": {}, |
||
| 1891 | "output_type": "execute_result" |
||
| 1892 | } |
||
| 1893 | ], |
||
| 1894 | "source": [ |
||
| 1895 | "# Are the number of predictions correct?\n", |
||
| 1896 | "# Should be 57458.\n", |
||
| 1897 | "\n", |
||
| 1898 | "len(predictions)" |
||
| 1899 | ] |
||
| 1900 | }, |
||
| 1901 | { |
||
| 1902 | "cell_type": "code", |
||
| 1903 | "execution_count": 53, |
||
| 1904 | "metadata": {}, |
||
| 1905 | "outputs": [ |
||
| 1906 | { |
||
| 1907 | "data": { |
||
| 1908 | "text/html": [ |
||
| 1909 | "<div>\n", |
||
| 1910 | "<style scoped>\n", |
||
| 1911 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 1912 | " vertical-align: middle;\n", |
||
| 1913 | " }\n", |
||
| 1914 | "\n", |
||
| 1915 | " .dataframe tbody tr th {\n", |
||
| 1916 | " vertical-align: top;\n", |
||
| 1917 | " }\n", |
||
| 1918 | "\n", |
||
| 1919 | " .dataframe thead th {\n", |
||
| 1920 | " text-align: right;\n", |
||
| 1921 | " }\n", |
||
| 1922 | "</style>\n", |
||
| 1923 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 1924 | " <thead>\n", |
||
| 1925 | " <tr style=\"text-align: right;\">\n", |
||
| 1926 | " <th></th>\n", |
||
| 1927 | " <th>no_tumor_tissue</th>\n", |
||
| 1928 | " <th>has_tumor_tissue</th>\n", |
||
| 1929 | " </tr>\n", |
||
| 1930 | " </thead>\n", |
||
| 1931 | " <tbody>\n", |
||
| 1932 | " <tr>\n", |
||
| 1933 | " <th>0</th>\n", |
||
| 1934 | " <td>0.477844</td>\n", |
||
| 1935 | " <td>0.522156</td>\n", |
||
| 1936 | " </tr>\n", |
||
| 1937 | " <tr>\n", |
||
| 1938 | " <th>1</th>\n", |
||
| 1939 | " <td>0.477844</td>\n", |
||
| 1940 | " <td>0.522156</td>\n", |
||
| 1941 | " </tr>\n", |
||
| 1942 | " <tr>\n", |
||
| 1943 | " <th>2</th>\n", |
||
| 1944 | " <td>0.477844</td>\n", |
||
| 1945 | " <td>0.522156</td>\n", |
||
| 1946 | " </tr>\n", |
||
| 1947 | " <tr>\n", |
||
| 1948 | " <th>3</th>\n", |
||
| 1949 | " <td>0.477844</td>\n", |
||
| 1950 | " <td>0.522156</td>\n", |
||
| 1951 | " </tr>\n", |
||
| 1952 | " <tr>\n", |
||
| 1953 | " <th>4</th>\n", |
||
| 1954 | " <td>0.477844</td>\n", |
||
| 1955 | " <td>0.522156</td>\n", |
||
| 1956 | " </tr>\n", |
||
| 1957 | " </tbody>\n", |
||
| 1958 | "</table>\n", |
||
| 1959 | "</div>" |
||
| 1960 | ], |
||
| 1961 | "text/plain": [ |
||
| 1962 | " no_tumor_tissue has_tumor_tissue\n", |
||
| 1963 | "0 0.477844 0.522156\n", |
||
| 1964 | "1 0.477844 0.522156\n", |
||
| 1965 | "2 0.477844 0.522156\n", |
||
| 1966 | "3 0.477844 0.522156\n", |
||
| 1967 | "4 0.477844 0.522156" |
||
| 1968 | ] |
||
| 1969 | }, |
||
| 1970 | "execution_count": 53, |
||
| 1971 | "metadata": {}, |
||
| 1972 | "output_type": "execute_result" |
||
| 1973 | } |
||
| 1974 | ], |
||
| 1975 | "source": [ |
||
| 1976 | "# Put the predictions into a dataframe\n", |
||
| 1977 | "\n", |
||
| 1978 | "df_preds = pd.DataFrame(predictions, columns=['no_tumor_tissue', 'has_tumor_tissue'])\n", |
||
| 1979 | "\n", |
||
| 1980 | "df_preds.head()" |
||
| 1981 | ] |
||
| 1982 | }, |
||
| 1983 | { |
||
| 1984 | "cell_type": "code", |
||
| 1985 | "execution_count": 54, |
||
| 1986 | "metadata": {}, |
||
| 1987 | "outputs": [ |
||
| 1988 | { |
||
| 1989 | "data": { |
||
| 1990 | "text/html": [ |
||
| 1991 | "<div>\n", |
||
| 1992 | "<style scoped>\n", |
||
| 1993 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 1994 | " vertical-align: middle;\n", |
||
| 1995 | " }\n", |
||
| 1996 | "\n", |
||
| 1997 | " .dataframe tbody tr th {\n", |
||
| 1998 | " vertical-align: top;\n", |
||
| 1999 | " }\n", |
||
| 2000 | "\n", |
||
| 2001 | " .dataframe thead th {\n", |
||
| 2002 | " text-align: right;\n", |
||
| 2003 | " }\n", |
||
| 2004 | "</style>\n", |
||
| 2005 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 2006 | " <thead>\n", |
||
| 2007 | " <tr style=\"text-align: right;\">\n", |
||
| 2008 | " <th></th>\n", |
||
| 2009 | " <th>no_tumor_tissue</th>\n", |
||
| 2010 | " <th>has_tumor_tissue</th>\n", |
||
| 2011 | " <th>file_names</th>\n", |
||
| 2012 | " </tr>\n", |
||
| 2013 | " </thead>\n", |
||
| 2014 | " <tbody>\n", |
||
| 2015 | " <tr>\n", |
||
| 2016 | " <th>0</th>\n", |
||
| 2017 | " <td>0.477844</td>\n", |
||
| 2018 | " <td>0.522156</td>\n", |
||
| 2019 | " <td>test_images/00006537328c33e284c973d7b39d340809...</td>\n", |
||
| 2020 | " </tr>\n", |
||
| 2021 | " <tr>\n", |
||
| 2022 | " <th>1</th>\n", |
||
| 2023 | " <td>0.477844</td>\n", |
||
| 2024 | " <td>0.522156</td>\n", |
||
| 2025 | " <td>test_images/0000ec92553fda4ce39889f9226ace43ca...</td>\n", |
||
| 2026 | " </tr>\n", |
||
| 2027 | " <tr>\n", |
||
| 2028 | " <th>2</th>\n", |
||
| 2029 | " <td>0.477844</td>\n", |
||
| 2030 | " <td>0.522156</td>\n", |
||
| 2031 | " <td>test_images/00024a6dee61f12f7856b0fc6be20bc7a4...</td>\n", |
||
| 2032 | " </tr>\n", |
||
| 2033 | " <tr>\n", |
||
| 2034 | " <th>3</th>\n", |
||
| 2035 | " <td>0.477844</td>\n", |
||
| 2036 | " <td>0.522156</td>\n", |
||
| 2037 | " <td>test_images/000253dfaa0be9d0d100283b22284ab2f6...</td>\n", |
||
| 2038 | " </tr>\n", |
||
| 2039 | " <tr>\n", |
||
| 2040 | " <th>4</th>\n", |
||
| 2041 | " <td>0.477844</td>\n", |
||
| 2042 | " <td>0.522156</td>\n", |
||
| 2043 | " <td>test_images/000270442cc15af719583a8172c87cd2bd...</td>\n", |
||
| 2044 | " </tr>\n", |
||
| 2045 | " </tbody>\n", |
||
| 2046 | "</table>\n", |
||
| 2047 | "</div>" |
||
| 2048 | ], |
||
| 2049 | "text/plain": [ |
||
| 2050 | " no_tumor_tissue has_tumor_tissue \\\n", |
||
| 2051 | "0 0.477844 0.522156 \n", |
||
| 2052 | "1 0.477844 0.522156 \n", |
||
| 2053 | "2 0.477844 0.522156 \n", |
||
| 2054 | "3 0.477844 0.522156 \n", |
||
| 2055 | "4 0.477844 0.522156 \n", |
||
| 2056 | "\n", |
||
| 2057 | " file_names \n", |
||
| 2058 | "0 test_images/00006537328c33e284c973d7b39d340809... \n", |
||
| 2059 | "1 test_images/0000ec92553fda4ce39889f9226ace43ca... \n", |
||
| 2060 | "2 test_images/00024a6dee61f12f7856b0fc6be20bc7a4... \n", |
||
| 2061 | "3 test_images/000253dfaa0be9d0d100283b22284ab2f6... \n", |
||
| 2062 | "4 test_images/000270442cc15af719583a8172c87cd2bd... " |
||
| 2063 | ] |
||
| 2064 | }, |
||
| 2065 | "execution_count": 54, |
||
| 2066 | "metadata": {}, |
||
| 2067 | "output_type": "execute_result" |
||
| 2068 | } |
||
| 2069 | ], |
||
| 2070 | "source": [ |
||
| 2071 | "# This outputs the file names in the sequence in which \n", |
||
| 2072 | "# the generator processed the test images.\n", |
||
| 2073 | "test_filenames = test_gen.filenames\n", |
||
| 2074 | "\n", |
||
| 2075 | "# add the filenames to the dataframe\n", |
||
| 2076 | "df_preds['file_names'] = test_filenames\n", |
||
| 2077 | "\n", |
||
| 2078 | "df_preds.head()" |
||
| 2079 | ] |
||
| 2080 | }, |
||
| 2081 | { |
||
| 2082 | "cell_type": "code", |
||
| 2083 | "execution_count": 55, |
||
| 2084 | "metadata": {}, |
||
| 2085 | "outputs": [ |
||
| 2086 | { |
||
| 2087 | "data": { |
||
| 2088 | "text/html": [ |
||
| 2089 | "<div>\n", |
||
| 2090 | "<style scoped>\n", |
||
| 2091 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 2092 | " vertical-align: middle;\n", |
||
| 2093 | " }\n", |
||
| 2094 | "\n", |
||
| 2095 | " .dataframe tbody tr th {\n", |
||
| 2096 | " vertical-align: top;\n", |
||
| 2097 | " }\n", |
||
| 2098 | "\n", |
||
| 2099 | " .dataframe thead th {\n", |
||
| 2100 | " text-align: right;\n", |
||
| 2101 | " }\n", |
||
| 2102 | "</style>\n", |
||
| 2103 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 2104 | " <thead>\n", |
||
| 2105 | " <tr style=\"text-align: right;\">\n", |
||
| 2106 | " <th></th>\n", |
||
| 2107 | " <th>no_tumor_tissue</th>\n", |
||
| 2108 | " <th>has_tumor_tissue</th>\n", |
||
| 2109 | " <th>file_names</th>\n", |
||
| 2110 | " <th>id</th>\n", |
||
| 2111 | " </tr>\n", |
||
| 2112 | " </thead>\n", |
||
| 2113 | " <tbody>\n", |
||
| 2114 | " <tr>\n", |
||
| 2115 | " <th>0</th>\n", |
||
| 2116 | " <td>0.477844</td>\n", |
||
| 2117 | " <td>0.522156</td>\n", |
||
| 2118 | " <td>test_images/00006537328c33e284c973d7b39d340809...</td>\n", |
||
| 2119 | " <td>00006537328c33e284c973d7b39d340809f7271b</td>\n", |
||
| 2120 | " </tr>\n", |
||
| 2121 | " <tr>\n", |
||
| 2122 | " <th>1</th>\n", |
||
| 2123 | " <td>0.477844</td>\n", |
||
| 2124 | " <td>0.522156</td>\n", |
||
| 2125 | " <td>test_images/0000ec92553fda4ce39889f9226ace43ca...</td>\n", |
||
| 2126 | " <td>0000ec92553fda4ce39889f9226ace43cae3364e</td>\n", |
||
| 2127 | " </tr>\n", |
||
| 2128 | " <tr>\n", |
||
| 2129 | " <th>2</th>\n", |
||
| 2130 | " <td>0.477844</td>\n", |
||
| 2131 | " <td>0.522156</td>\n", |
||
| 2132 | " <td>test_images/00024a6dee61f12f7856b0fc6be20bc7a4...</td>\n", |
||
| 2133 | " <td>00024a6dee61f12f7856b0fc6be20bc7a48ba3d2</td>\n", |
||
| 2134 | " </tr>\n", |
||
| 2135 | " <tr>\n", |
||
| 2136 | " <th>3</th>\n", |
||
| 2137 | " <td>0.477844</td>\n", |
||
| 2138 | " <td>0.522156</td>\n", |
||
| 2139 | " <td>test_images/000253dfaa0be9d0d100283b22284ab2f6...</td>\n", |
||
| 2140 | " <td>000253dfaa0be9d0d100283b22284ab2f6b643f6</td>\n", |
||
| 2141 | " </tr>\n", |
||
| 2142 | " <tr>\n", |
||
| 2143 | " <th>4</th>\n", |
||
| 2144 | " <td>0.477844</td>\n", |
||
| 2145 | " <td>0.522156</td>\n", |
||
| 2146 | " <td>test_images/000270442cc15af719583a8172c87cd2bd...</td>\n", |
||
| 2147 | " <td>000270442cc15af719583a8172c87cd2bd9c7746</td>\n", |
||
| 2148 | " </tr>\n", |
||
| 2149 | " </tbody>\n", |
||
| 2150 | "</table>\n", |
||
| 2151 | "</div>" |
||
| 2152 | ], |
||
| 2153 | "text/plain": [ |
||
| 2154 | " no_tumor_tissue has_tumor_tissue \\\n", |
||
| 2155 | "0 0.477844 0.522156 \n", |
||
| 2156 | "1 0.477844 0.522156 \n", |
||
| 2157 | "2 0.477844 0.522156 \n", |
||
| 2158 | "3 0.477844 0.522156 \n", |
||
| 2159 | "4 0.477844 0.522156 \n", |
||
| 2160 | "\n", |
||
| 2161 | " file_names \\\n", |
||
| 2162 | "0 test_images/00006537328c33e284c973d7b39d340809... \n", |
||
| 2163 | "1 test_images/0000ec92553fda4ce39889f9226ace43ca... \n", |
||
| 2164 | "2 test_images/00024a6dee61f12f7856b0fc6be20bc7a4... \n", |
||
| 2165 | "3 test_images/000253dfaa0be9d0d100283b22284ab2f6... \n", |
||
| 2166 | "4 test_images/000270442cc15af719583a8172c87cd2bd... \n", |
||
| 2167 | "\n", |
||
| 2168 | " id \n", |
||
| 2169 | "0 00006537328c33e284c973d7b39d340809f7271b \n", |
||
| 2170 | "1 0000ec92553fda4ce39889f9226ace43cae3364e \n", |
||
| 2171 | "2 00024a6dee61f12f7856b0fc6be20bc7a48ba3d2 \n", |
||
| 2172 | "3 000253dfaa0be9d0d100283b22284ab2f6b643f6 \n", |
||
| 2173 | "4 000270442cc15af719583a8172c87cd2bd9c7746 " |
||
| 2174 | ] |
||
| 2175 | }, |
||
| 2176 | "execution_count": 55, |
||
| 2177 | "metadata": {}, |
||
| 2178 | "output_type": "execute_result" |
||
| 2179 | } |
||
| 2180 | ], |
||
| 2181 | "source": [ |
||
| 2182 | "# Create an id column\n", |
||
| 2183 | "\n", |
||
| 2184 | "# A file name now has this format: \n", |
||
| 2185 | "# test_images/00006537328c33e284c973d7b39d340809f7271b.tif\n", |
||
| 2186 | "\n", |
||
| 2187 | "# This function will extract the id:\n", |
||
| 2188 | "# 00006537328c33e284c973d7b39d340809f7271b\n", |
||
| 2189 | "\n", |
||
| 2190 | "\n", |
||
| 2191 | "def extract_id(x):\n", |
||
| 2192 | " \n", |
||
| 2193 | " # split into a list\n", |
||
| 2194 | " a = x.split('/')\n", |
||
| 2195 | " # split into a list\n", |
||
| 2196 | " b = a[1].split('.')\n", |
||
| 2197 | " extracted_id = b[0]\n", |
||
| 2198 | " \n", |
||
| 2199 | " return extracted_id\n", |
||
| 2200 | "\n", |
||
| 2201 | "df_preds['id'] = df_preds['file_names'].apply(extract_id)\n", |
||
| 2202 | "\n", |
||
| 2203 | "df_preds.head()" |
||
| 2204 | ] |
||
| 2205 | }, |
||
| 2206 | { |
||
| 2207 | "cell_type": "code", |
||
| 2208 | "execution_count": 56, |
||
| 2209 | "metadata": {}, |
||
| 2210 | "outputs": [], |
||
| 2211 | "source": [ |
||
| 2212 | "# Get the predicted labels.\n", |
||
| 2213 | "# We were asked to predict a probability that the image has tumor tissue\n", |
||
| 2214 | "y_pred = df_preds['has_tumor_tissue']\n", |
||
| 2215 | "\n", |
||
| 2216 | "# get the id column\n", |
||
| 2217 | "image_id = df_preds['id']" |
||
| 2218 | ] |
||
| 2219 | }, |
||
| 2220 | { |
||
| 2221 | "cell_type": "markdown", |
||
| 2222 | "metadata": {}, |
||
| 2223 | "source": [ |
||
| 2224 | "### Confusion Matrix" |
||
| 2225 | ] |
||
| 2226 | }, |
||
| 2227 | { |
||
| 2228 | "cell_type": "markdown", |
||
| 2229 | "metadata": {}, |
||
| 2230 | "source": [ |
||
| 2231 | "# 5. Submission" |
||
| 2232 | ] |
||
| 2233 | }, |
||
| 2234 | { |
||
| 2235 | "cell_type": "code", |
||
| 2236 | "execution_count": 57, |
||
| 2237 | "metadata": {}, |
||
| 2238 | "outputs": [ |
||
| 2239 | { |
||
| 2240 | "data": { |
||
| 2241 | "text/html": [ |
||
| 2242 | "<div>\n", |
||
| 2243 | "<style scoped>\n", |
||
| 2244 | " .dataframe tbody tr th:only-of-type {\n", |
||
| 2245 | " vertical-align: middle;\n", |
||
| 2246 | " }\n", |
||
| 2247 | "\n", |
||
| 2248 | " .dataframe tbody tr th {\n", |
||
| 2249 | " vertical-align: top;\n", |
||
| 2250 | " }\n", |
||
| 2251 | "\n", |
||
| 2252 | " .dataframe thead th {\n", |
||
| 2253 | " text-align: right;\n", |
||
| 2254 | " }\n", |
||
| 2255 | "</style>\n", |
||
| 2256 | "<table border=\"1\" class=\"dataframe\">\n", |
||
| 2257 | " <thead>\n", |
||
| 2258 | " <tr style=\"text-align: right;\">\n", |
||
| 2259 | " <th></th>\n", |
||
| 2260 | " <th>label</th>\n", |
||
| 2261 | " </tr>\n", |
||
| 2262 | " <tr>\n", |
||
| 2263 | " <th>id</th>\n", |
||
| 2264 | " <th></th>\n", |
||
| 2265 | " </tr>\n", |
||
| 2266 | " </thead>\n", |
||
| 2267 | " <tbody>\n", |
||
| 2268 | " <tr>\n", |
||
| 2269 | " <th>00006537328c33e284c973d7b39d340809f7271b</th>\n", |
||
| 2270 | " <td>0.522156</td>\n", |
||
| 2271 | " </tr>\n", |
||
| 2272 | " <tr>\n", |
||
| 2273 | " <th>0000ec92553fda4ce39889f9226ace43cae3364e</th>\n", |
||
| 2274 | " <td>0.522156</td>\n", |
||
| 2275 | " </tr>\n", |
||
| 2276 | " <tr>\n", |
||
| 2277 | " <th>00024a6dee61f12f7856b0fc6be20bc7a48ba3d2</th>\n", |
||
| 2278 | " <td>0.522156</td>\n", |
||
| 2279 | " </tr>\n", |
||
| 2280 | " <tr>\n", |
||
| 2281 | " <th>000253dfaa0be9d0d100283b22284ab2f6b643f6</th>\n", |
||
| 2282 | " <td>0.522156</td>\n", |
||
| 2283 | " </tr>\n", |
||
| 2284 | " <tr>\n", |
||
| 2285 | " <th>000270442cc15af719583a8172c87cd2bd9c7746</th>\n", |
||
| 2286 | " <td>0.522156</td>\n", |
||
| 2287 | " </tr>\n", |
||
| 2288 | " </tbody>\n", |
||
| 2289 | "</table>\n", |
||
| 2290 | "</div>" |
||
| 2291 | ], |
||
| 2292 | "text/plain": [ |
||
| 2293 | " label\n", |
||
| 2294 | "id \n", |
||
| 2295 | "00006537328c33e284c973d7b39d340809f7271b 0.522156\n", |
||
| 2296 | "0000ec92553fda4ce39889f9226ace43cae3364e 0.522156\n", |
||
| 2297 | "00024a6dee61f12f7856b0fc6be20bc7a48ba3d2 0.522156\n", |
||
| 2298 | "000253dfaa0be9d0d100283b22284ab2f6b643f6 0.522156\n", |
||
| 2299 | "000270442cc15af719583a8172c87cd2bd9c7746 0.522156" |
||
| 2300 | ] |
||
| 2301 | }, |
||
| 2302 | "execution_count": 57, |
||
| 2303 | "metadata": {}, |
||
| 2304 | "output_type": "execute_result" |
||
| 2305 | } |
||
| 2306 | ], |
||
| 2307 | "source": [ |
||
| 2308 | "submission = pd.DataFrame({'id':image_id, \n", |
||
| 2309 | " 'label':y_pred, \n", |
||
| 2310 | " }).set_index('id')\n", |
||
| 2311 | "\n", |
||
| 2312 | "submission.to_csv('patch_preds.csv', columns=['label']) \n", |
||
| 2313 | "submission.head()" |
||
| 2314 | ] |
||
| 2315 | }, |
||
| 2316 | { |
||
| 2317 | "cell_type": "code", |
||
| 2318 | "execution_count": 58, |
||
| 2319 | "metadata": {}, |
||
| 2320 | "outputs": [], |
||
| 2321 | "source": [ |
||
| 2322 | "# Delete the test_dir directory we created to prevent a Kaggle error.\n", |
||
| 2323 | "# Kaggle allows a max of 500 files to be saved.\n", |
||
| 2324 | "\n", |
||
| 2325 | "shutil.rmtree('test_dir')" |
||
| 2326 | ] |
||
| 2327 | }, |
||
| 2328 | { |
||
| 2329 | "cell_type": "markdown", |
||
| 2330 | "metadata": {}, |
||
| 2331 | "source": [ |
||
| 2332 | "# Confusion Matrix" |
||
| 2333 | ] |
||
| 2334 | }, |
||
| 2335 | { |
||
| 2336 | "cell_type": "code", |
||
| 2337 | "execution_count": 59, |
||
| 2338 | "metadata": {}, |
||
| 2339 | "outputs": [ |
||
| 2340 | { |
||
| 2341 | "data": { |
||
| 2342 | "image/png": "iVBORw0KGgoAAAANSUhEUgAAAiQAAAHFCAYAAADCA+LKAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjcuMCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy88F64QAAAACXBIWXMAAA9hAAAPYQGoP6dpAABRUElEQVR4nO3deVxUZfs/8M/IMizJKCAzoKhoqCjmgopQpgW4JKJPpRZJLogapk1qGvkU2AJBpaa4oJmYS+hT4WNmfMVKcgFFlFJCW8TtkREXxA1Zz+8Pf54aQWcG53Rw+ryf13m9mnuuOecaXk9xcd33uY9CEAQBRERERDJqIncCRERERCxIiIiISHYsSIiIiEh2LEiIiIhIdixIiIiISHYsSIiIiEh2LEiIiIhIdixIiIiISHYsSIiIiEh2LEjIov38888YP348vLy8YGdnh4ceegg9e/ZEUlISLl26JOm1Dx06hP79+0OlUkGhUGDhwoVmv4ZCoUBcXJzZz2tIamoqFAoFFAoFdu7cWed9QRDw8MMPQ6FQYMCAAQ26xtKlS5GammrSZ3bu3HnXnIiocbOWOwEiqaxcuRLR0dHo2LEjXnvtNXTu3BlVVVU4cOAAli9fjuzsbKSnp0t2/QkTJuD69etIS0tD8+bN0bZtW7NfIzs7G61atTL7eY3VtGlTrFq1qk7RkZWVhT/++ANNmzZt8LmXLl0KV1dXjBs3zujP9OzZE9nZ2ejcuXODr0tE8mBBQhYpOzsbL730EkJCQrB582YolUrxvZCQEMycORMZGRmS5nDkyBFERUVhyJAhkl2jb9++kp3bGKNHj8b69euxZMkSODk5ieOrVq1CQEAArly58rfkUVVVBYVCAScnJ9l/JkTUMJyyIYsUHx8PhUKBFStW6BUjt9na2iIsLEx8XVtbi6SkJHTq1AlKpRJubm548cUXcebMGb3PDRgwAL6+vsjNzUW/fv3g4OCAdu3a4f3330dtbS2AP6czqqursWzZMnFqAwDi4uLEf/6r2585ceKEOPb9999jwIABcHFxgb29PVq3bo1nnnkGN27cEGPqm7I5cuQIhg8fjubNm8POzg7du3fHmjVr9GJuT218/vnnmDt3Ljw8PODk5ITg4GAcO3bMuB8ygOeffx4A8Pnnn4tjZWVl+PLLLzFhwoR6PzNv3jz4+/vD2dkZTk5O6NmzJ1atWoW/Puezbdu2KCgoQFZWlvjzu91hup372rVrMXPmTLRs2RJKpRK///57nSmbCxcuwNPTE4GBgaiqqhLP/8svv8DR0RERERFGf1cikhYLErI4NTU1+P777+Hn5wdPT0+jPvPSSy9hzpw5CAkJwZYtW/DOO+8gIyMDgYGBuHDhgl6sTqfDCy+8gDFjxmDLli0YMmQIYmJisG7dOgDA0KFDkZ2dDQB49tlnkZ2dLb421okTJzB06FDY2tri008/RUZGBt5//304OjqisrLyrp87duwYAgMDUVBQgEWLFuGrr75C586dMW7cOCQlJdWJf+ONN3Dy5El88sknWLFiBX777TcMGzYMNTU1RuXp5OSEZ599Fp9++qk49vnnn6NJkyYYPXr0Xb/b5MmTsWnTJnz11Vd4+umnMW3aNLzzzjtiTHp6Otq1a4cePXqIP787p9diYmJw6tQpLF++HF9//TXc3NzqXMvV1RVpaWnIzc3FnDlzAAA3btzAyJEj0bp1ayxfvtyo70lEfwOByMLodDoBgPDcc88ZFV9YWCgAEKKjo/XG9+3bJwAQ3njjDXGsf//+AgBh3759erGdO3cWBg0apDcGQJg6dareWGxsrFDfv3arV68WAAhFRUWCIAjCF198IQAQ8vPz75k7ACE2NlZ8/dxzzwlKpVI4deqUXtyQIUMEBwcH4fLly4IgCMIPP/wgABCeeuopvbhNmzYJAITs7Ox7Xvd2vrm5ueK5jhw5IgiCIPTu3VsYN26cIAiC0KVLF6F///53PU9NTY1QVVUlvP3224KLi4tQW1srvne3z96+3uOPP37X93744Qe98cTERAGAkJ6eLowdO1awt7cXfv7553t+RyL6e7FDQv94P/zwAwDUWTzZp08f+Pj44LvvvtMb12g06NOnj97YI488gpMnT5otp+7du8PW1haTJk3CmjVrcPz4caM+9/333yMoKKhOZ2jcuHG4ceNGnU7NX6etgFvfA4BJ36V///5o3749Pv30Uxw+fBi5ubl3na65nWNwcDBUKhWsrKxgY2ODt956CxcvXkRJSYnR133mmWeMjn3ttdcwdOhQPP/881izZg0WL16Mrl27Gv15IpIeCxKyOK6urnBwcEBRUZFR8RcvXgQAuLu713nPw8NDfP82FxeXOnFKpRLl5eUNyLZ+7du3x44dO+Dm5oapU6eiffv2aN++PT7++ON7fu7ixYt3/R633/+rO7/L7fU2pnwXhUKB8ePHY926dVi+fDk6dOiAfv361Ru7f/9+DBw4EMCtu6D27NmD3NxczJ071+Tr1vc975XjuHHjcPPmTWg0Gq4dIWqEWJCQxbGyskJQUBDy8vLqLEqtz+1fysXFxXXeO3v2LFxdXc2Wm52dHQCgoqJCb/zOdSoA0K9fP3z99dcoKytDTk4OAgICoNVqkZaWdtfzu7i43PV7ADDrd/mrcePG4cKFC1i+fDnGjx9/17i0tDTY2Nhg69atGDVqFAIDA9GrV68GXbO+xcF3U1xcjKlTp6J79+64ePEiZs2a1aBrEpF0WJCQRYqJiYEgCIiKiqp3EWhVVRW+/vprAMCTTz4JAOKi1Ntyc3NRWFiIoKAgs+V1+06Rn3/+WW/8di71sbKygr+/P5YsWQIAOHjw4F1jg4KC8P3334sFyG2fffYZHBwcJLsltmXLlnjttdcwbNgwjB079q5xCoUC1tbWsLKyEsfKy8uxdu3aOrHm6jrV1NTg+eefh0KhwLfffouEhAQsXrwYX3311X2fm4jMh/uQkEUKCAjAsmXLEB0dDT8/P7z00kvo0qULqqqqcOjQIaxYsQK+vr4YNmwYOnbsiEmTJmHx4sVo0qQJhgwZghMnTuDNN9+Ep6cnXn31VbPl9dRTT8HZ2RmRkZF4++23YW1tjdTUVJw+fVovbvny5fj+++8xdOhQtG7dGjdv3hTvZAkODr7r+WNjY7F161Y88cQTeOutt+Ds7Iz169fjm2++QVJSElQqldm+y53ef/99gzFDhw7F/PnzER4ejkmTJuHixYv48MMP6701u2vXrkhLS8PGjRvRrl072NnZNWjdR2xsLHbt2oXt27dDo9Fg5syZyMrKQmRkJHr06AEvLy+Tz0lE5seChCxWVFQU+vTpgwULFiAxMRE6nQ42Njbo0KEDwsPD8fLLL4uxy5YtQ/v27bFq1SosWbIEKpUKgwcPRkJCQr1rRhrKyckJGRkZ0Gq1GDNmDJo1a4aJEydiyJAhmDhxohjXvXt3bN++HbGxsdDpdHjooYfg6+uLLVu2iGsw6tOxY0fs3bsXb7zxBqZOnYry8nL4+Phg9erVJu14KpUnn3wSn376KRITEzFs2DC0bNkSUVFRcHNzQ2RkpF7svHnzUFxcjKioKFy9ehVt2rTR26fFGJmZmUhISMCbb76p1+lKTU1Fjx49MHr0aOzevRu2trbm+HpEdB8UgvCX3YiIiIiIZMA1JERERCQ7FiREREQkOxYkREREJDsWJERERBaouroa//73v+Hl5QV7e3u0a9cOb7/9tvggUAAQBAFxcXHw8PCAvb09BgwYgIKCAr3zVFRUYNq0aXB1dYWjoyPCwsLq7PFUWlqKiIgIqFQqqFQqRERE4PLlyybly4KEiIjIAiUmJmL58uVITk5GYWEhkpKS8MEHH2Dx4sViTFJSEubPn4/k5GTk5uZCo9EgJCQEV69eFWO0Wi3S09ORlpaG3bt349q1awgNDdV7CGd4eDjy8/ORkZGBjIwM5Ofnm7wjMu+yISIiskChoaFQq9VYtWqVOPbMM8/AwcEBa9euhSAI8PDwgFarFZ+GXVFRAbVajcTEREyePBllZWVo0aIF1q5dKz7B++zZs/D09MS2bdswaNAgFBYWonPnzsjJyYG/vz8AiLtLHz16FB07djQqX3ZIiIiIHhAVFRW4cuWK3nHnoyhue+yxx/Ddd9/h119/BQD89NNP2L17N5566ikAQFFREXQ6nd7eRkqlEv3798fevXsBAHl5eaiqqtKL8fDwgK+vrxiTnZ0NlUolFiMA0LdvX6hUKjHGGBa5MZq1bUu5UyBqlMrP7pI7BaJGx8a1neTXqLpg3BO7DUlI/gzz5s3TG4uNjUVcXFyd2Dlz5qCsrAydOnWClZUVampq8N577+H5558HAOh0OgCAWq3W+5xarRaf+K3T6WBra4vmzZvXibn9eZ1OBzc3tzrXd3NzE2OMYZEFCRERkSWKiYnBjBkz9Mbqe/QCAGzcuBHr1q3Dhg0b0KVLF+Tn50Or1cLDw0PvmVN3PqhSEASDD6+8M6a+eGPO81csSIiIiKRWW2M4xghKpfKuBcidXnvtNbz++ut47rnnANx6PtTJkyeRkJCAsWPHQqPRALjV4XB3dxc/V1JSInZNNBoNKisrUVpaqtclKSkpQWBgoBhz7ty5Otc/f/58ne7LvXANCRERkdSEWvMcJrhx4waaNNH/NW9lZSXe9uvl5QWNRoPMzEzx/crKSmRlZYnFhp+fH2xsbPRiiouLceTIETEmICAAZWVl2L9/vxizb98+lJWViTHGYIeEiIhIarWmFRPmMGzYMLz33nto3bo1unTpgkOHDmH+/PmYMGECgFvTLFqtFvHx8fD29oa3tzfi4+Ph4OCA8PBwAIBKpUJkZCRmzpwJFxcXODs7Y9asWejatav45HEfHx8MHjwYUVFRSElJAQBMmjQJoaGhRt9hA7AgISIiskiLFy/Gm2++iejoaJSUlMDDwwOTJ0/GW2+9JcbMnj0b5eXliI6ORmlpKfz9/bF9+3Y0bdpUjFmwYAGsra0xatQolJeXIygoCKmpqbCyshJj1q9fj+nTp4t344SFhSE5OdmkfC1yHxLeZUNUP95lQ1TX33GXTeXZAsNBRrD16GKW8zRG7JAQERFJTYYpmwcNF7USERGR7NghISIikpqJd8j8E7EgISIikpqZ9iGxZJyyISIiItmxQ0JERCQ1TtkYxIKEiIhIarzLxiBO2RAREZHs2CEhIiKSmMApG4NYkBAREUmNUzYGsSAhIiKSGjskBnENCREREcmOHRIiIiKpcWM0g1iQEBERSY1TNgZxyoaIiIhkxw4JERGR1HiXjUEsSIiIiKTGKRuDOGVDREREsmOHhIiISGqcsjGIBQkREZHEBIG3/RrCKRsiIiKSHTskREREUuOiVoNYkBAREUmNa0gMYkFCREQkNXZIDOIaEiIiIpIdOyRERERS48P1DGJBQkREJDVO2RjEKRsiIiKSHTskREREUuNdNgaxICEiIpIap2wM4pQNERERyY4dEiIiIqlxysYgFiRERERSY0FiEKdsiIiISHbskBAREUlMELgxmiHskBAREUmtttY8hwnatm0LhUJR55g6dSoAQBAExMXFwcPDA/b29hgwYAAKCgr0zlFRUYFp06bB1dUVjo6OCAsLw5kzZ/RiSktLERERAZVKBZVKhYiICFy+fNnkHxELEiIiIqkJteY5TJCbm4vi4mLxyMzMBACMHDkSAJCUlIT58+cjOTkZubm50Gg0CAkJwdWrV8VzaLVapKenIy0tDbt378a1a9cQGhqKmpo/Oz7h4eHIz89HRkYGMjIykJ+fj4iICJN/RApBEASTP9XIWdu2lDsFokap/OwuuVMganRsXNtJfo3yHz4xy3nsn5jY4M9qtVps3boVv/32GwDAw8MDWq0Wc+bMAXCrG6JWq5GYmIjJkyejrKwMLVq0wNq1azF69GgAwNmzZ+Hp6Ylt27Zh0KBBKCwsROfOnZGTkwN/f38AQE5ODgICAnD06FF07NjR6PzYISEiIpKamaZsKioqcOXKFb2joqLC4OUrKyuxbt06TJgwAQqFAkVFRdDpdBg4cKAYo1Qq0b9/f+zduxcAkJeXh6qqKr0YDw8P+Pr6ijHZ2dlQqVRiMQIAffv2hUqlEmOMxYKEiIhIamaasklISBDXatw+EhISDF5+8+bNuHz5MsaNGwcA0Ol0AAC1Wq0Xp1arxfd0Oh1sbW3RvHnze8a4ubnVuZ6bm5sYYyzeZUNERPSAiImJwYwZM/TGlEqlwc+tWrUKQ4YMgYeHh964QqHQey0IQp2xO90ZU1+8Mee5EwsSIiIiqZlpYzSlUmlUAfJXJ0+exI4dO/DVV1+JYxqNBsCtDoe7u7s4XlJSInZNNBoNKisrUVpaqtclKSkpQWBgoBhz7ty5Otc8f/58ne6LIZyyISIikpoMd9nctnr1ari5uWHo0KHimJeXFzQajXjnDXBrnUlWVpZYbPj5+cHGxkYvpri4GEeOHBFjAgICUFZWhv3794sx+/btQ1lZmRhjLHZIiIiILFRtbS1Wr16NsWPHwtr6z1/5CoUCWq0W8fHx8Pb2hre3N+Lj4+Hg4IDw8HAAgEqlQmRkJGbOnAkXFxc4Oztj1qxZ6Nq1K4KDgwEAPj4+GDx4MKKiopCSkgIAmDRpEkJDQ026wwZgQUJERCQ9mZ5ls2PHDpw6dQoTJkyo897s2bNRXl6O6OholJaWwt/fH9u3b0fTpk3FmAULFsDa2hqjRo1CeXk5goKCkJqaCisrKzFm/fr1mD59ung3TlhYGJKTk03OlfuQEP2DcB8Sorr+ln1IvllolvPYD9Wa5TyNEdeQEBERkew4ZUNERCS1Bi5I/SdhQUJERCQ1mdaQPEhYkBAREUmNHRKDuIaEiIiIZMcOCRERkdQ4ZWMQCxIiIiKpccrGIE7ZEBERkezYISEiIpIap2wMYkFCREQkNRYkBnHKhoiIiGTHDgkREZHULO+xcWbHgoSIiEhqnLIxiFM2REREJDt2SIiIiKTGDolBLEiIiIikxo3RDGJBQkREJDV2SAziGhIiIiKSHTskREREUuNtvwaxICEiIpIap2wM4pQNERERyY4dEiIiIqmxQ2IQCxIiIiKp8bZfgzhlQ0RERLJjh4SIiEhiQi3vsjGEBQkREZHUuIbEIE7ZEBERkezYISEiIpIaF7UaxIKEiIhIalxDYhALEiIiIqlxDYlBXENCREREsmOHhIiISGrskBjEgoSIiEhqfNqvQZyyISIiItmxICHJTJk8Fr8dy8a1K39gX863eOzRPnKnRGSS6uoaLFqxBoOeHQe/J4Zj8MjxWPbpetRK3H7P/GE3wl6YhB4DhiHshUnYkbVH7/209K3414svwT/kafiHPI0XJr2KXdm5kuZE96m21jyHBWNBQpIYOTIM8z+KQ8L7i9CrzyDs3r0fW79eB09PD7lTIzLaqvWbsGnzNrwxIxpbNqzAjOgJWL3hS6z/YkuDz7n5m0yMe3n2Xd/PP1KIWbEJGDYoCF+uWYphg4Iw680E/FxwVIzRtHDFq1PGY+OqRdi4ahH6+HXDtNffxu/HTzY4L5JYrWCew0T/+9//MGbMGLi4uMDBwQHdu3dHXl6e+L4gCIiLi4OHhwfs7e0xYMAAFBQU6J2joqIC06ZNg6urKxwdHREWFoYzZ87oxZSWliIiIgIqlQoqlQoRERG4fPmySbmyICFJvPpKFD5dnYZPV3+Oo0d/x8xZsTh95iymTH5R7tSIjPbTkaN4ol9f9A/sg5buagx8oh8C+/REwdHfxJiqqip8tGQVnhw+Br2DRuD5KC32H/y5wddcu3EzAnr3RNSLo9GujSeiXhwN/17dsXbTZjFmwGN98XhgH7Rt3QptW7fCK5PHwcHeDj/9pWghKi0txaOPPgobGxt8++23+OWXX/DRRx+hWbNmYkxSUhLmz5+P5ORk5ObmQqPRICQkBFevXhVjtFot0tPTkZaWht27d+PatWsIDQ1FTU2NGBMeHo78/HxkZGQgIyMD+fn5iIiIMClfWRe1njlzBsuWLcPevXuh0+mgUCigVqsRGBiIKVOmwNPTU870qIFsbGzQs+cjSPxgid54ZmYWAvr2kikrItP1fKQLNm3+BidOnUHb1q1w9LfjOPhzAV5/ZbIY8+/35uN/uhJ8MO91tHB1xnc/7sWUmf9G+mfL0MazpcnX/KmgEC+O+pfe2KN9/PQKkr+qqanB//2wC+U3b6K7byeTr0d/Exl2ak1MTISnpydWr14tjrVt2/bPlAQBCxcuxNy5c/H0008DANasWQO1Wo0NGzZg8uTJKCsrw6pVq7B27VoEBwcDANatWwdPT0/s2LEDgwYNQmFhITIyMpCTkwN/f38AwMqVKxEQEIBjx46hY8eORuUrW4dk9+7d8PHxQXp6Orp164YXX3wRY8aMQbdu3bB582Z06dIFe/bsMXwianRcXZ1hbW2NknMX9MZLSi5ArXGTKSsi00WOGYkhwQMwLHwSuj8eipHjX0bEqBF4KmQAAODUmbPYtiML8995A37dfdG6lQfGhz+Lno90Qfo3mQ265oWLpXBxbqY35uLcDBcuXdIb+/WPIvQO/hd6PhGGdz5Ixsfxb6K9V5sGXZP+BjJM2WzZsgW9evXCyJEj4ebmhh49emDlypXi+0VFRdDpdBg4cKA4plQq0b9/f+zduxcAkJeXh6qqKr0YDw8P+Pr6ijHZ2dlQqVRiMQIAffv2hUqlEmOMIVuH5NVXX8XEiROxYMGCu76v1WqRm3vvhVoVFRWoqKjQGxMEAQqFwmy5UsMId9zmplAo6owRNWbffpeFrdu/R2LcbDzs1QZHfzuOxI9T4ObqjOFPhaDw1z8gCAKGPj9R73NVlVVQOTkBAIp1JQgb82dHpaamBtXVNegd/GcXJHTgk4idPU18fed/v+r7b5pX61b4MnUJrly9hsydezD3vY+QmpzEosTC1fc7T6lUQqlU1ok9fvw4li1bhhkzZuCNN97A/v37MX36dCiVSrz44ovQ6XQAALVarfc5tVqNkydvrUfS6XSwtbVF8+bN68Tc/rxOp4ObW90/Nt3c3MQYY8hWkBw5cgTr1q276/uTJ0/G8uXLDZ4nISEB8+bN0xtTNHkICiun+86RGubChUuorq6GWtNCb7xFCxeUnDsvU1ZEpvtoySpMHDMKTwUPAAB0aO+FYl0JPlm7CcOfCkFtbS2srJpg06rFsLLSbzg72NsBAFq4uuDL1D+nL3dk7UHmzj1IjP1zYaujo4P4z64uzXHhYqneuS6VlsHljl8INjY2aN3q1iJxX58OKDj6K9b957+InT39/r84mZ1gpjtk6vudFxsbi7i4uDqxtbW16NWrF+Lj4wEAPXr0QEFBAZYtW4YXX/xzPZ8xBfCd7oypL97U5oBsUzbu7u73bOVkZ2fD3d3d4HliYmJQVlamdyiaNDVnqmSiqqoqHDz4M4KDHtcbDw5+HNk5B2TKish0N29WQNFE/z+oTZo0Qe3/7/T5dGiPmppaXCq9jNatPPQOVxdnAIC1tZXeuHOzZlAqbfXGXJo3E8/frYsPsnMP6l1zb+5BdO/qc89cBUFAZWWVGb41ScJMUzb1/c6LiYmp95Lu7u7o3Lmz3piPjw9OnToFANBoNABQp4tRUlIidk00Gg0qKytRWlp6z5hz587Vuf758+frdF/uRbYOyaxZszBlyhTk5eUhJCQEarUaCoUCOp0OmZmZ+OSTT7Bw4UKD56mvVcXpGvkt+Hgl1qz+GHl5PyFnXx6iIsegtWdLpKxYK3dqREYb8Kg/Vq5Jg7vaDQ97tUHhr7/js41f4V9Db82nt23dCkMHPoE33v0Qs16Ogk+H9igtK8P+vJ/g3a4tHg80fe+dMaOGY9zU17Bq3SY80S8AP+zKRk7uIXy27EMxZuHyVPTr2wsadQtcv3ED3+7IQu6hw1j+0Ttm++5kZmZa1Hq36Zn6PProozh27Jje2K+//oo2bW5N63l5eUGj0SAzMxM9evQAAFRWViIrKwuJiYkAAD8/P9jY2CAzMxOjRo0CABQXF+PIkSNISkoCAAQEBKCsrAz79+9Hnz63/j+/b98+lJWVITAw0OjvJltBEh0dDRcXFyxYsAApKSni7UNWVlbw8/PDZ599Jn55evD85z9b4OLcHP+e+yrc3d1wpOAYhoVF4NSp/8mdGpHR3nj1JSxe+Rne/XAJLpVeRgtXZ4wc/hReGh8uxrw7dwZSUj/Hh8krce78RTRTNUW3Lj7oF9C7Qdfs0bUzPpj3Ohav+AyLV66FZ0t3fPB2DB7p8ucdNBdLSxHzzgc4f/ESmjo6osPDXlj+0TsI7NPzvr8zWY5XX30VgYGBiI+Px6hRo7B//36sWLECK1asAHDrj3etVov4+Hh4e3vD29sb8fHxcHBwQHj4rf+Pq1QqREZGYubMmXBxcYGzszNmzZqFrl27infd+Pj4YPDgwYiKikJKSgoAYNKkSQgNDTX6DhsAUAiNYJVhVVUVLly4dUeGq6srbGxs7ut81ram32pH9E9QfnaX3CkQNTo2ru0kv8b1t18wy3kc31pvUvzWrVsRExOD3377DV5eXpgxYwaioqLE9wVBwLx585CSkoLS0lL4+/tjyZIl8PX1FWNu3ryJ1157DRs2bEB5eTmCgoKwdOlSva05Ll26hOnTp2PLllubBoaFhSE5OVlvzxNDGkVBYm4sSIjqx4KEqK6/pSCJe94s53GM+9ws52mMuFMrERERyU7WnVqJiIj+ERrwHJp/GhYkREREUpNh6/gHDadsiIiISHbskBAREUmNUzYGsSAhIiKSmLm2jrdknLIhIiIi2bFDQkREJDVO2RjEgoSIiEhqLEgMYkFCREQkNd72axDXkBAREZHs2CEhIiKSGqdsDGJBQkREJDGBBYlBnLIhIiIi2bFDQkREJDV2SAxiQUJERCQ17tRqEKdsiIiISHbskBAREUmNUzYGsSAhIiKSGgsSgzhlQ0RERLJjh4SIiEhigsAOiSEsSIiIiKTGKRuDWJAQERFJjQWJQVxDQkRERLJjh4SIiEhifJaNYSxIiIiIpMaCxCBO2RAREZHs2CEhIiKSGh9lYxALEiIiIolxDYlhnLIhIiIi2bFDQkREJDV2SAxiQUJERCQ1riExiFM2REREJDt2SIiIiCTGRa2GsSAhIiKSGqdsDOKUDRERkcSEWsEshyni4uKgUCj0Do1G82dOgoC4uDh4eHjA3t4eAwYMQEFBgd45KioqMG3aNLi6usLR0RFhYWE4c+aMXkxpaSkiIiKgUqmgUqkQERGBy5cvm/wzYkFCRERkobp06YLi4mLxOHz4sPheUlIS5s+fj+TkZOTm5kKj0SAkJARXr14VY7RaLdLT05GWlobdu3fj2rVrCA0NRU1NjRgTHh6O/Px8ZGRkICMjA/n5+YiIiDA5V07ZEBERSU2mKRtra2u9rshtgiBg4cKFmDt3Lp5++mkAwJo1a6BWq7FhwwZMnjwZZWVlWLVqFdauXYvg4GAAwLp16+Dp6YkdO3Zg0KBBKCwsREZGBnJycuDv7w8AWLlyJQICAnDs2DF07NjR6FzZISEiIpKYUGuew1S//fYbPDw84OXlheeeew7Hjx8HABQVFUGn02HgwIFirFKpRP/+/bF3714AQF5eHqqqqvRiPDw84OvrK8ZkZ2dDpVKJxQgA9O3bFyqVSowxFjskRERED4iKigpUVFTojSmVSiiVyjqx/v7++Oyzz9ChQwecO3cO7777LgIDA1FQUACdTgcAUKvVep9Rq9U4efIkAECn08HW1hbNmzevE3P78zqdDm5ubnWu7ebmJsYYix0SIiIiqdWa50hISBAXj94+EhIS6r3kkCFD8Mwzz6Br164IDg7GN998A+DW1MxtCoVC7zOCINQZu9OdMfXFG3OeO7EgISIikpi5pmxiYmJQVlamd8TExBiVg6OjI7p27YrffvtNXFdyZxejpKRE7JpoNBpUVlaitLT0njHnzp2rc63z58/X6b4YwoKEiIjoAaFUKuHk5KR31DddU5+KigoUFhbC3d0dXl5e0Gg0yMzMFN+vrKxEVlYWAgMDAQB+fn6wsbHRiykuLsaRI0fEmICAAJSVlWH//v1izL59+1BWVibGGItrSIiIiKQmw102s2bNwrBhw9C6dWuUlJTg3XffxZUrVzB27FgoFApotVrEx8fD29sb3t7eiI+Ph4ODA8LDwwEAKpUKkZGRmDlzJlxcXODs7IxZs2aJU0AA4OPjg8GDByMqKgopKSkAgEmTJiE0NNSkO2wAFiRERESSa8gdMvfrzJkzeP7553HhwgW0aNECffv2RU5ODtq0aQMAmD17NsrLyxEdHY3S0lL4+/tj+/btaNq0qXiOBQsWwNraGqNGjUJ5eTmCgoKQmpoKKysrMWb9+vWYPn26eDdOWFgYkpOTTc5XIQiCxW2wb23bUu4UiBql8rO75E6BqNGxcW0n+TVKgvqb5Txu32WZ5TyNEdeQEBERkew4ZUNERCQxOaZsHjQsSIiIiKQmmLYnxz8Rp2yIiIhIdvddkNTU1CA/P7/OxilERER0i1zPsnmQmFyQaLVarFq1CsCtYqR///7o2bMnPD09sXPnTnPnR0RE9MATahVmOSyZyQXJF198gW7dugEAvv76axQVFeHo0aPQarWYO3eu2RMkIiIiy2dyQXLhwgVxD/xt27Zh5MiR6NChAyIjI3H48GGzJ0hERPSg45SNYSYXJGq1Gr/88gtqamqQkZEhbh9748YNvZ3biIiI6BZBUJjlsGQm3/Y7fvx4jBo1Cu7u7lAoFAgJCQFw62E6nTp1MnuCREREZPlMLkji4uLg6+uL06dPY+TIkeJTBq2srPD666+bPUEiIqIHnaVPt5gDn2VD9A/CZ9kQ1fV3PMvmdO8gs5zHM/c7s5ynMTKqQ7Jo0SKjTzh9+vQGJ0NERGSJLO9Pf/MzqiBZsGCBUSdTKBQsSIiIiMhkRhUkRUVFUudBRERksSx9UzNzaPDW8ZWVlTh27Biqq6vNmQ8REZHF4U6thplckNy4cQORkZFwcHBAly5dcOrUKQC31o68//77Zk+QiIiILJ/JBUlMTAx++ukn7Ny5E3Z2duJ4cHAwNm7caNbkiIiILIEgmOewZCbvQ7J582Zs3LgRffv2hULxZ/uoc+fO+OOPP8yaHBERkSWw9OkWczC5Q3L+/Hm4ubnVGb9+/bpegUJERERkLJMLkt69e+Obb74RX98uQlauXImAgADzZUZERGQh+Cwbw0yesklISMDgwYPxyy+/oLq6Gh9//DEKCgqQnZ2NrKwsKXIkIiJ6oHHreMNM7pAEBgZiz549uHHjBtq3b4/t27dDrVYjOzsbfn5+UuRIREREFs7kDgkAdO3aFWvWrDF3LkRERBap1sKnW8yhQQVJTU0N0tPTUVhYCIVCAR8fHwwfPhzW1g06HRERkUWz9PUf5mByBXHkyBEMHz4cOp0OHTt2BAD8+uuvaNGiBbZs2YKuXbuaPUkiIqIHGW/7NczkNSQTJ05Ely5dcObMGRw8eBAHDx7E6dOn8cgjj2DSpElS5EhEREQWzuQOyU8//YQDBw6gefPm4ljz5s3x3nvvoXfv3mZNjoiIyBJY+i6r5mByh6Rjx444d+5cnfGSkhI8/PDDZkmKiIjIkvDheoYZVZBcuXJFPOLj4zF9+nR88cUXOHPmDM6cOYMvvvgCWq0WiYmJUudLREREFkghCIYbSU2aNNHbFv72R26P/fV1TU2NFHmaxNq2pdwpEDVK5Wd3yZ0CUaNj49pO8mscaRdqlvP4Ht9qlvM0RkatIfnhhx+kzoOIiMhi8bZfw4wqSPr37y91HkRERPQP1uCdzG7cuIFTp06hsrJSb/yRRx6576SIiIgsCe+yMczkguT8+fMYP348vv3223rfbwxrSIiIiBoTbh1vmMm3/Wq1WpSWliInJwf29vbIyMjAmjVr4O3tjS1btkiRIxEREVk4kwuS77//HgsWLEDv3r3RpEkTtGnTBmPGjEFSUhISEhKkyJGIiOiBJggKsxz3IyEhAQqFAlqt9i95CYiLi4OHhwfs7e0xYMAAFBQU6H2uoqIC06ZNg6urKxwdHREWFoYzZ87oxZSWliIiIgIqlQoqlQoRERG4fPmySfmZXJBcv34dbm5uAABnZ2ecP38ewK0nAB88eNDU0xEREVk8QTDP0VC5ublYsWJFnXWeSUlJmD9/PpKTk5GbmwuNRoOQkBBcvXpVjNFqtUhPT0daWhp2796Na9euITQ0VG+JRnh4OPLz85GRkYGMjAzk5+cjIiLCpBwbtFPrsWPHAADdu3dHSkoK/ve//2H58uVwd3c39XREREQWr1ZQmOVoiGvXruGFF17AypUr9R77IggCFi5ciLlz5+Lpp5+Gr68v1qxZgxs3bmDDhg0AgLKyMqxatQofffQRgoOD0aNHD6xbtw6HDx/Gjh07AACFhYXIyMjAJ598goCAAAQEBGDlypXYunWrWC8Yo0FrSIqLiwEAsbGxyMjIQOvWrbFo0SLEx8ebejoiIiIyUkVFhd7u6VeuXEFFRcU9PzN16lQMHToUwcHBeuNFRUXQ6XQYOHCgOKZUKtG/f3/s3bsXAJCXl4eqqiq9GA8PD/j6+oox2dnZUKlU8Pf3F2P69u0LlUolxhjD5LtsXnjhBfGfe/TogRMnTuDo0aNo3bo1XF1dTT0dERGRxTPXxmgJCQmYN2+e3lhsbCzi4uLqjU9LS8PBgweRm5tb5z2dTgcAUKvVeuNqtRonT54UY2xtbfU6K7djbn9ep9OJSzn+ys3NTYwxRoP3IbnNwcEBPXv2vN/TEBERWSxz3fYbExODGTNm6I0plcp6Y0+fPo1XXnkF27dvh52d3V3P+ddHwwC3pnLuHLvTnTH1xRtznr8yqiC588vfy/z5842OJSIiIuMplcq7FiB3ysvLQ0lJCfz8/MSxmpoa/Pjjj0hOThbXd+h0Or01oCUlJWLXRKPRoLKyEqWlpXpdkpKSEgQGBoox586dq3P98+fP1+m+3ItRBcmhQ4eMOpkplRAREdE/hRwbtQYFBeHw4cN6Y+PHj0enTp0wZ84ctGvXDhqNBpmZmejRowcAoLKyEllZWUhMTAQA+Pn5wcbGBpmZmRg1ahQAoLi4GEeOHEFSUhIAICAgAGVlZdi/fz/69OkDANi3bx/KysrEosUYfLgeERGRxOTYqbVp06bw9fXVG3N0dISLi4s4rtVqER8fD29vb3h7eyM+Ph4ODg4IDw8HAKhUKkRGRmLmzJlwcXGBs7MzZs2aha5du4qLZH18fDB48GBERUUhJSUFADBp0iSEhoaiY8eORud732tIiIiI6ME0e/ZslJeXIzo6GqWlpfD398f27dvRtGlTMWbBggWwtrbGqFGjUF5ejqCgIKSmpsLKykqMWb9+PaZPny7ejRMWFobk5GSTclEIguU98sfatqXcKRA1SuVnd8mdAlGjY+PaTvJr7NE8a5bzPKr7wiznaYzYISEiIpJYrdwJPABM3hiNiIiIyNzYISEiIpKYAN6FakiDOiRr167Fo48+Cg8PD3E3t4ULF+K///2vWZMjIiKyBLWCeQ5LZnJBsmzZMsyYMQNPPfUULl++LD7tr1mzZli4cKG58yMiInrg1UJhlsOSmVyQLF68GCtXrsTcuXP1bvnp1atXnQ1YiIiIiIxh8hqSoqIicUe3v1Iqlbh+/bpZkiIiIrIkXENimMkdEi8vL+Tn59cZ//bbb9G5c2dz5ERERGRRas10WDKTOySvvfYapk6dips3b0IQBOzfvx+ff/45EhIS8Mknn0iRIxEREVk4kwuS8ePHo7q6GrNnz8aNGzcQHh6Oli1b4uOPP8Zzzz0nRY5EREQPNE7ZGHZfW8dfuHABtbW1cHNzM2dO941bxxPVj1vHE9X1d2wdn6E2zx/sg8+lmeU8jdF9bYzm6upqrjyIiIjoH8zkgsTLywsKxd1bT8ePH7+vhIiIiCyNpS9INQeTCxKtVqv3uqqqCocOHUJGRgZee+01c+VFRERkMbiGxDCTC5JXXnml3vElS5bgwIED950QERER/fOY7Wm/Q4YMwZdffmmu0xEREVmMWoV5Dktmtqf9fvHFF3B2djbX6YiIiCyGpT+HxhxMLkh69Oiht6hVEATodDqcP38eS5cuNWtyRERElsDCH9RrFiYXJCNGjNB73aRJE7Ro0QIDBgxAp06dzJUXERER/YOYVJBUV1ejbdu2GDRoEDQajVQ5ERERWRTe9muYSYtara2t8dJLL6GiokKqfIiIiCxOrUJhlsOSmXyXjb+/Pw4dOiRFLkRERPQPZfIakujoaMycORNnzpyBn58fHB0d9d5/5JFHzJYcERGRJeCiVsOMLkgmTJiAhQsXYvTo0QCA6dOni+8pFAoIggCFQoGamhrzZ0lERPQA4xoSw4wuSNasWYP3338fRUVFUuZDRERE/0BGFySCcKvh1KZNG8mSISIiskSWvsuqOZi0huReT/klIiKi+nGnVsNMKkg6dOhgsCi5dOnSfSVERERE/zwmFSTz5s2DSqWSKhciIiKLxLtsDDOpIHnuuefg5uYmVS5EREQWiWtIDDO6IOH6ESIioobhbb+GGb1T6+27bIiIiIjMzegOSW0t6zsiIqKG4J/0hpm8dTwRERGZhmtIDDP54XpERERE5saChIiISGK1ZjpMsWzZMjzyyCNwcnKCk5MTAgIC8O2334rvC4KAuLg4eHh4wN7eHgMGDEBBQYHeOSoqKjBt2jS4urrC0dERYWFhOHPmjF5MaWkpIiIioFKpoFKpEBERgcuXL5uYLQsSIiIiyclRkLRq1Qrvv/8+Dhw4gAMHDuDJJ5/E8OHDxaIjKSkJ8+fPR3JyMnJzc6HRaBASEoKrV6+K59BqtUhPT0daWhp2796Na9euITQ0VO9BuuHh4cjPz0dGRgYyMjKQn5+PiIgIk39GCsECb5+xtm0pdwpEjVL52V1yp0DU6Ni4tpP8GimtxpjlPJPPrLuvzzs7O+ODDz7AhAkT4OHhAa1Wizlz5gC41Q1Rq9VITEzE5MmTUVZWhhYtWmDt2rUYPXo0AODs2bPw9PTEtm3bMGjQIBQWFqJz587IycmBv78/ACAnJwcBAQE4evQoOnbsaHRu7JAQERFJTFCY56ioqMCVK1f0joqKCoPXr6mpQVpaGq5fv46AgAAUFRVBp9Nh4MCBYoxSqUT//v2xd+9eAEBeXh6qqqr0Yjw8PODr6yvGZGdnQ6VSicUIAPTt2xcqlUqMMRYLEiIiIomZa8omISFBXKtx+0hISLjrdQ8fPoyHHnoISqUSU6ZMQXp6Ojp37gydTgcAUKvVevFqtVp8T6fTwdbWFs2bN79nTH07uLu5uYkxxuJtv0RERA+ImJgYzJgxQ29MqVTeNb5jx47Iz8/H5cuX8eWXX2Ls2LHIysoS379zF3ZBEAzuzH5nTH3xxpznTixIiIiIJGaurUWVSuU9C5A72dra4uGHHwYA9OrVC7m5ufj444/FdSM6nQ7u7u5ifElJidg10Wg0qKysRGlpqV6XpKSkBIGBgWLMuXPn6lz3/PnzdbovhnDKhoiISGKCmY77zkMQUFFRAS8vL2g0GmRmZorvVVZWIisrSyw2/Pz8YGNjoxdTXFyMI0eOiDEBAQEoKyvD/v37xZh9+/ahrKxMjDEWOyREREQSk2On1jfeeANDhgyBp6cnrl69irS0NOzcuRMZGRlQKBTQarWIj4+Ht7c3vL29ER8fDwcHB4SHhwMAVCoVIiMjMXPmTLi4uMDZ2RmzZs1C165dERwcDADw8fHB4MGDERUVhZSUFADApEmTEBoaatIdNgALEiIiIot07tw5REREoLi4GCqVCo888ggyMjIQEhICAJg9ezbKy8sRHR2N0tJS+Pv7Y/v27WjatKl4jgULFsDa2hqjRo1CeXk5goKCkJqaCisrKzFm/fr1mD59ung3TlhYGJKTk03Ol/uQEP2DcB8Sorr+jn1IFrQ2zz4kr566v31IGjN2SIiIiCRmrkWtloyLWomIiEh27JAQERFJzOLWRkiABQkREZHE5LjL5kHDKRsiIiKSHTskREREEuOiVsNYkBAREUmMa0gM45QNERERyY4dEiIiIonVskdiEAsSIiIiiXENiWEsSIiIiCTG/ohhXENCREREsmOHhIiISGKcsjGMBQkREZHEuFOrYZyyISIiItmxQ0JERCQx3vZrGAsSIiIiibEcMYxTNkRERCQ7dkiIiIgkxrtsDGNBQkREJDGuITGMUzZEREQkO3ZIiIiIJMb+iGEsSIiIiCTGNSSGsSAhIiKSGNeQGMY1JERERCQ7dkiIiIgkxv6IYSxIiIiIJMY1JIZxyoaIiIhkxw4JERGRxARO2hjEgoSIiEhinLIxjFM2REREJDt2SIiIiCTGfUgMY0FCREQkMZYjhnHKhoiIiGTHgoQkM2XyWPx2LBvXrvyBfTnf4rFH+8idEpFJqqtrsGjFGgx6dhz8nhiOwSPHY9mn61FbK+0SxcwfdiPshUnoMWAYwl6YhB1Ze/TeT0vfin+9+BL8Q56Gf8jTeGHSq9iVnStpTnR/aiGY5bBkLEhIEiNHhmH+R3FIeH8RevUZhN2792Pr1+vg6ekhd2pERlu1fhM2bd6GN2ZEY8uGFZgRPQGrN3yJ9V9safA5N3+TiXEvz77r+/lHCjErNgHDBgXhyzVLMWxQEGa9mYCfC46KMZoWrnh1ynhsXLUIG1ctQh+/bpj2+tv4/fjJBudF0qo102GKhIQE9O7dG02bNoWbmxtGjBiBY8eO6cUIgoC4uDh4eHjA3t4eAwYMQEFBgV5MRUUFpk2bBldXVzg6OiIsLAxnzpzRiyktLUVERARUKhVUKhUiIiJw+fJlk/JlQUKSePWVKHy6Og2frv4cR4/+jpmzYnH6zFlMmfyi3KkRGe2nI0fxRL++6B/YBy3d1Rj4RD8E9umJgqO/iTFVVVX4aMkqPDl8DHoHjcDzUVrsP/hzg6+5duNmBPTuiagXR6NdG09EvTga/r26Y+2mzWLMgMf64vHAPmjbuhXatm6FVyaPg4O9HX76S9FCjYtgpv+ZIisrC1OnTkVOTg4yMzNRXV2NgQMH4vr162JMUlIS5s+fj+TkZOTm5kKj0SAkJARXr14VY7RaLdLT05GWlobdu3fj2rVrCA0NRU1NjRgTHh6O/Px8ZGRkICMjA/n5+YiIiDApXxYkZHY2Njbo2fMRZO7I0hvPzMxCQN9eMmVFZLqej3TBvgP5OHHq1l+DR387joM/F+DxgN5izL/fm49Dh3/BB/Nex5drlmLgE49hysx/4+Tp/zXomj8VFCKwd0+9sUf7+CH/cGG98TU1Ndi2YyfKb95Ed99ODbomWaaMjAyMGzcOXbp0Qbdu3bB69WqcOnUKeXl5AG51RxYuXIi5c+fi6aefhq+vL9asWYMbN25gw4YNAICysjKsWrUKH330EYKDg9GjRw+sW7cOhw8fxo4dOwAAhYWFyMjIwCeffIKAgAAEBARg5cqV2Lp1a52OzL006rtsTp8+jdjYWHz66ad3jamoqEBFRYXemCAIUCgUUqdHd+Hq6gxra2uUnLugN15ScgFqjZtMWRGZLnLMSFy9dh3DwifBqkkT1NTWYvqksXgqZAAA4NSZs9i2Iwvfpa+FWwsXAMD48GexZ18e0r/JhHbKOJOveeFiKVycm+mNuTg3w4VLl/TGfv2jCC9MnoHKyko42Nvj4/g30d6rTUO+Jv0NzLXqqL7feUqlEkql0uBny8rKAADOzs4AgKKiIuh0OgwcOFDvXP3798fevXsxefJk5OXloaqqSi/Gw8MDvr6+2Lt3LwYNGoTs7GyoVCr4+/uLMX379oVKpcLevXvRsWNHo75boy5ILl26hDVr1tyzIElISMC8efP0xhRNHoLCyknq9MgAQdBvLyoUijpjRI3Zt99lYev275EYNxsPe7XB0d+OI/HjFLi5OmP4UyEo/PUPCIKAoc9P1PtcVWUVVE63/htUrCtB2JjJ4ns1NTWorq5B7+B/iWOhA59E7Oxp4us7/6Cq748sr9at8GXqEly5eg2ZO/dg7nsfITU5iUVJI2WurePr+50XGxuLuLi4e19fEDBjxgw89thj8PX1BQDodDoAgFqt1otVq9U4efKkGGNra4vmzZvXibn9eZ1OBze3un9surm5iTHGkLUg2bLl3gvDjh8/bvAcMTExmDFjht5Ycxe2LeV04cIlVFdXQ61poTfeooULSs6dlykrItN9tGQVJo4ZhaeCBwAAOrT3QrGuBJ+s3YThT4WgtrYWVlZNsGnVYlhZ6c+AO9jbAQBauLrgy9Ql4viOrD3I3LkHibF/Lmx1dHQQ/9nVpTkuXCzVO9el0jK43PELwcbGBq1b3Vok7uvTAQVHf8W6//wXsbOn3/8Xp0arvt95xnRHXn75Zfz888/YvXt3nfeMKYDvdGdMffGmzlbIWpCMGDHC4F/Nhr5Mfa0qTtfIq6qqCgcP/ozgoMfx3/9miOPBwY/j66//T8bMiExz82YFFE30/3vSpEkT1P7//2b5dGiPmppaXCq9DL/uvvWew9raSiwcAMC5WTMolbZ6Y3/VrYsPsnMP4sXn/uyg7M09iO5dfe6ZqyAIqKysMup70d/PXFM2xk7P/NW0adOwZcsW/Pjjj2jVqpU4rtFoANzqcLi7u4vjJSUlYtdEo9GgsrISpaWlel2SkpISBAYGijHnzp2rc93z58/X6b7ci6yLWt3d3fHll1+itra23uPgwYNypkf3YcHHKxE54XmMGzsanTo9jI8+iENrz5ZIWbFW7tSIjDbgUX+sXJOGrL378b/ic9iRtQefbfwKQY8HAADatm6FoQOfwBvvfojMnXtw5qwOhwuPYdW6Tfhx7/4GXXPMqOHYm3sQq9ZtwvGTp7Fq3Sbk5B5CxKgRYszC5anIyz+C/xWfw69/FOHjlFTkHjqMoQOfMMfXJgnUCoJZDlMIgoCXX34ZX331Fb7//nt4eXnpve/l5QWNRoPMzExxrLKyEllZWWKx4efnBxsbG72Y4uJiHDlyRIwJCAhAWVkZ9u//8//z+/btQ1lZmRhjDFk7JH5+fjh48CBGjBhR7/tcc/Dg+s9/tsDFuTn+PfdVuLu74UjBMQwLi8CpUw2784BIDm+8+hIWr/wM7364BJdKL6OFqzNGDn8KL40PF2PenTsDKamf48PklTh3/iKaqZqiWxcf9PvLnTim6NG1Mz6Y9zoWr/gMi1euhWdLd3zwdgwe6fLnVPTF0lLEvPMBzl+8hKaOjujwsBeWf/QOAvv0vMeZ6Z9m6tSp2LBhA/773/+iadOm4noOlUoFe3t7KBQKaLVaxMfHw9vbG97e3oiPj4eDgwPCw8PF2MjISMycORMuLi5wdnbGrFmz0LVrVwQHBwMAfHx8MHjwYERFRSElJQUAMGnSJISGhhq9oBUAFIKMv/F37dqF69evY/DgwfW+f/36dRw4cAD9+/c36bzWti3NkR6RxSk/u0vuFIgaHRvXdpJfY0ybp81ynnUnvzI69m7LF1avXo1x48YBuNVFmTdvHlJSUlBaWgp/f38sWbJEXPgKADdv3sRrr72GDRs2oLy8HEFBQVi6dCk8PT3FmEuXLmH69Oni2tCwsDAkJyejWbNmxucrZ0EiFRYkRPVjQUJU199RkIS3+ZfhICNsOJlulvM0RtwYjYiIiGTXqPchISIisgTm2ofEkrEgISIikpi0z4e2DCxIiIiIJFbLDolBXENCREREsmOHhIiISGJcQ2IYCxIiIiKJcQ2JYZyyISIiItmxQ0JERCQxC9yD1OxYkBAREUmMd9kYxikbIiIikh07JERERBLjolbDWJAQERFJjLf9GsYpGyIiIpIdOyREREQS46JWw1iQEBERSYy3/RrGgoSIiEhiXNRqGNeQEBERkezYISEiIpIY77IxjAUJERGRxLio1TBO2RAREZHs2CEhIiKSGO+yMYwFCRERkcQ4ZWMYp2yIiIhIduyQEBERSYx32RjGgoSIiEhitVxDYhCnbIiIiEh27JAQERFJjP0Rw1iQEBERSYx32RjGgoSIiEhiLEgM4xoSIiIikh07JERERBLjTq2GsSAhIiKSGKdsDOOUDREREcmOHRIiIiKJcadWw9ghISIikpggCGY5TPXjjz9i2LBh8PDwgEKhwObNm+vkFRcXBw8PD9jb22PAgAEoKCjQi6moqMC0adPg6uoKR0dHhIWF4cyZM3oxpaWliIiIgEqlgkqlQkREBC5fvmxSrixIiIiILNT169fRrVs3JCcn1/t+UlIS5s+fj+TkZOTm5kKj0SAkJARXr14VY7RaLdLT05GWlobdu3fj2rVrCA0NRU1NjRgTHh6O/Px8ZGRkICMjA/n5+YiIiDApV4VggUt/rW1byp0CUaNUfnaX3CkQNTo2ru0kv0ZP98fMcp6Dxbsb/FmFQoH09HSMGDECwK3uiIeHB7RaLebMmQPgVjdErVYjMTERkydPRllZGVq0aIG1a9di9OjRAICzZ8/C09MT27Ztw6BBg1BYWIjOnTsjJycH/v7+AICcnBwEBATg6NGj6Nixo1H5sUNCREQkMbmmbO6lqKgIOp0OAwcOFMeUSiX69++PvXv3AgDy8vJQVVWlF+Ph4QFfX18xJjs7GyqVSixGAKBv375QqVRijDG4qJWIiOgBUVFRgYqKCr0xpVIJpVJp8rl0Oh0AQK1W642r1WqcPHlSjLG1tUXz5s3rxNz+vE6ng5ubW53zu7m5iTHGYIeEiIhIYrUQzHIkJCSIC0dvHwkJCfeVm0Kh0HstCEKdsTvdGVNfvDHn+SsWJERERBITzPS/mJgYlJWV6R0xMTENykmj0QBAnS5GSUmJ2DXRaDSorKxEaWnpPWPOnTtX5/znz5+v0325FxYkREREEqsVBLMcSqUSTk5OekdDpmsAwMvLCxqNBpmZmeJYZWUlsrKyEBgYCADw8/ODjY2NXkxxcTGOHDkixgQEBKCsrAz79+8XY/bt24eysjIxxhhcQ0JERGShrl27ht9//118XVRUhPz8fDg7O6N169bQarWIj4+Ht7c3vL29ER8fDwcHB4SHhwMAVCoVIiMjMXPmTLi4uMDZ2RmzZs1C165dERwcDADw8fHB4MGDERUVhZSUFADApEmTEBoaavQdNgALEiIiIsnJtVPrgQMH8MQTT4ivZ8yYAQAYO3YsUlNTMXv2bJSXlyM6OhqlpaXw9/fH9u3b0bRpU/EzCxYsgLW1NUaNGoXy8nIEBQUhNTUVVlZWYsz69esxffp08W6csLCwu+59cjfch4ToH4T7kBDV9XfsQ+Lj1scs5yks2W846AHFNSREREQkO07ZEBERSYwP1zOMBQkREZHEai1vdYTZccqGiIiIZMcOCRERkcQ4ZWMYCxIiIiKJccrGME7ZEBERkezYISEiIpIYp2wMY0FCREQkMUGolTuFRo8FCRERkcRq2SExiGtIiIiISHbskBAREUnMAh8bZ3YsSIiIiCTGKRvDOGVDREREsmOHhIiISGKcsjGMBQkREZHEuFOrYZyyISIiItmxQ0JERCQx7tRqGAsSIiIiiXENiWGcsiEiIiLZsUNCREQkMe5DYhgLEiIiIolxysYwFiREREQS422/hnENCREREcmOHRIiIiKJccrGMBYkREREEuOiVsM4ZUNERESyY4eEiIhIYpyyMYwFCRERkcR4l41hnLIhIiIi2bFDQkREJDE+XM8wFiREREQS45SNYZyyISIiItmxQ0JERCQx3mVjGAsSIiIiiXENiWGcsiEiIpKYIAhmORpi6dKl8PLygp2dHfz8/LBr1y4zfzvzYEFCRERkoTZu3AitVou5c+fi0KFD6NevH4YMGYJTp07JnVodCsECJ7asbVvKnQJRo1R+tnH+ZUQkJxvXdtJfw0y/l6oq/2dSvL+/P3r27Illy5aJYz4+PhgxYgQSEhLMkpO5sENCREQkMcFMhykqKyuRl5eHgQMH6o0PHDgQe/fubfB3kQoXtRIRET0gKioqUFFRoTemVCqhVCrrxF64cAE1NTVQq9V642q1GjqdTtI8G8IiC5JqE1taJI2KigokJCQgJiam3n9ZiP6p+O/GP4+5fi/FxcVh3rx5emOxsbGIi4u762cUCoXea0EQ6ow1Bha5hoQahytXrkClUqGsrAxOTk5yp0PUaPDfDWooUzoklZWVcHBwwH/+8x/861//EsdfeeUV5OfnIysrS/J8TcE1JERERA8IpVIJJycnveNuXTZbW1v4+fkhMzNTbzwzMxOBgYF/R7omscgpGyIiIgJmzJiBiIgI9OrVCwEBAVixYgVOnTqFKVOmyJ1aHSxIiIiILNTo0aNx8eJFvP322yguLoavry+2bduGNm3ayJ1aHSxISDJKpRKxsbFctEd0B/67QX+n6OhoREdHy52GQVzUSkRERLLjolYiIiKSHQsSIiIikh0LEiIiIpIdCxIiIiKSHQsSkszSpUvh5eUFOzs7+Pn5YdcuPmmW/tl+/PFHDBs2DB4eHlAoFNi8ebPcKRE1GixISBIbN26EVqvF3LlzcejQIfTr1w9DhgzBqVOn5E6NSDbXr19Ht27dkJycLHcqRI0Ob/slSfj7+6Nnz55YtmyZOObj44MRI0YgISFBxsyIGgeFQoH09HSMGDFC7lSIGgV2SMjsKisrkZeXh4EDB+qNDxw4EHv37pUpKyIiasxYkJDZXbhwATU1NVCr1XrjarUaOp1OpqyIiKgxY0FCklEoFHqvBUGoM0ZERASwICEJuLq6wsrKqk43pKSkpE7XhIiICGBBQhKwtbWFn58fMjMz9cYzMzMRGBgoU1ZERNSY8Wm/JIkZM2YgIiICvXr1QkBAAFasWIFTp05hypQpcqdGJJtr167h999/F18XFRUhPz8fzs7OaN26tYyZEcmPt/2SZJYuXYqkpCQUFxfD19cXCxYswOOPPy53WkSy2blzJ5544ok642PHjkVqaurfnxBRI8KChIiIiGTHNSREREQkOxYkREREJDsWJERERCQ7FiREREQkOxYkREREJDsWJERERCQ7FiREREQkOxYkRDKKi4tD9+7dxdfjxo3DiBEj/vY8Tpw4AYVCgfz8/LvGtG3bFgsXLjT6nKmpqWjWrNl956ZQKLB58+b7Pg8RNW4sSIjuMG7cOCgUCigUCtjY2KBdu3aYNWsWrl+/Lvm1P/74Y6N37DSmiCAielDwWTZE9Rg8eDBWr16Nqqoq7Nq1CxMnTsT169exbNmyOrFVVVWwsbExy3VVKpVZzkNE9KBhh4SoHkqlEhqNBp6enggPD8cLL7wgThvcnmb59NNP0a5dOyiVSgiCgLKyMkyaNAlubm5wcnLCk08+iZ9++knvvO+//z7UajWaNm2KyMhI3Lx5U+/9O6dsamtrkZiYiIcffhhKpRKtW7fGe++9BwDw8vICAPTo0QMKhQIDBgwQP7d69Wr4+PjAzs4OnTp1wtKlS/Wus3//fvTo0QN2dnbo1asXDh06ZPLPaP78+ejatSscHR3h6emJ6OhoXLt2rU7c5s2b0aFDB9jZ2SEkJASnT5/We//rr7+Gn58f7Ozs0K5dO8ybNw/V1dX1XrOyshIvv/wy3N3dYWdnh7Zt2yIhIcHk3Imo8WGHhMgI9vb2qKqqEl///vvv2LRpE7788ktYWVkBAIYOHQpnZ2ds27YNKpUKKSkpCAoKwq+//gpnZ2ds2rQJsbGxWLJkCfr164e1a9di0aJFaNeu3V2vGxMTg5UrV2LBggV47LHHUFxcjKNHjwK4VVT06dMHO3bsQJcuXWBrawsAWLlyJWJjY5GcnIwePXrg0KFDiIqKgqOjI8aOHYvr168jNDQUTz75JNatW4eioiK88sorJv9MmjRpgkWLFqFt27YoKipCdHQ0Zs+erVf83LhxA++99x7WrFkDW1tbREdH47nnnsOePXsAAP/3f/+HMWPGYNGiRejXrx/++OMPTJo0CQAQGxtb55qLFi3Cli1bsGnTJrRu3RqnT5+uU+AQ0QNKICI9Y8eOFYYPHy6+3rdvn+Di4iKMGjVKEARBiI2NFWxsbISSkhIx5rvvvhOcnJyEmzdv6p2rffv2QkpKiiAIghAQECBMmTJF731/f3+hW7du9V77ypUrglKpFFauXFlvnkVFRQIA4dChQ3rjnp6ewoYNG/TG3nnnHSEgIEAQBEFISUkRnJ2dhevXr4vvL1u2rN5z/VWbNm2EBQsW3PX9TZs2CS4uLuLr1atXCwCEnJwccaywsFAAIOzbt08QBEHo16+fEB8fr3eetWvXCu7u7uJrAEJ6erogCIIwbdo04cknnxRqa2vvmgcRPZjYISGqx9atW/HQQw+huroaVVVVGD58OBYvXiy+36ZNG7Ro0UJ8nZeXh2vXrsHFxUXvPOXl5fjjjz8AAIWFhZgyZYre+wEBAfjhhx/qzaGwsBAVFRUICgoyOu/z58/j9OnTiIyMRFRUlDheXV0trk8pLCxEt27d4ODgoJeHqX744QfEx8fjl19+wZUrV1BdXY2bN2/i+vXrcHR0BABYW1ujV69e4mc6deqEZs2aobCwEH369EFeXh5yc3PFaSgAqKmpwc2bN3Hjxg29HIFbU1ohISHo2LEjBg8ejNDQUAwcONDk3Imo8WFBQlSPJ554AsuWLYONjQ08PDzqLFq9/Qv3ttraWri7u2Pnzp11ztXQW1/t7e1N/kxtbS2AW9M2/v7+eu/dnloSBKFB+fzVyZMn8dRTT2HKlCl455134OzsjN27dyMyMlJvagu4ddvunW6P1dbWYt68eXj66afrxNjZ2dUZ69mzJ4qKivDtt99ix44dGDVqFIKDg/HFF1/c93ciInmxICGqh6OjIx5++GGj43v27AmdTgdra2u0bdu23hgfHx/k5OTgxRdfFMdycnLuek5vb2/Y29vju+++w8SJE+u8f3vNSE1NjTimVqvRsmVLHD9+HC+88EK95+3cuTPWrl2L8vJysei5Vx71OXDgAKqrq/HRRx+hSZNba+M3bdpUJ666uhoHDhxAnz59AADHjh3D5cuX0alTJwC3fm7Hjh0z6Wft5OSE0aNHY/To0Xj22WcxePBgXLp0Cc7OziZ9ByJqXFiQEJlBcHAwAgICMGLECCQmJqJjx444e/Ystm3bhhEjRqBXr1545ZVXMHbsWPTq1QuPPfYY1q9fj4KCgrsuarWzs8OcOXMwe/Zs2Nra4tFHH8X58+dRUFCAyMhIuLm5wd7eHhkZGWjVqhXs7OygUqkQFxeH6dOnw8nJCUOGDEFFRQUOHDiA0tJSzJgxA+Hh4Zg7dy4iIyPx73//GydOnMCHH35o0vdt3749qqursXjxYgwbNgx79uzB8uXL68TZ2Nhg2rRpWLRoEWxsbPDyyy+jb9++YoHy1ltvITQ0FJ6enhg5ciSaNGmCn3/+GYcPH8a7775b53wLFiyAu7s7unfvjiZNmuA///kPNBqNWTZgIyJ58bZfIjNQKBTYtm0bHn/8cUyYMAEdOnTAc889hxMnTkCtVgMARo8ejbfeegtz5syBn58fTp48iZdeeume533zzTcxc+ZMvPXWW/Dx8cHo0aNRUlIC4Nb6jEWLFiElJQUeHh4YPnw4AGDixIn45JNPkJqaiq5du6J///5ITU0VbxN+6KGH8PXXX+OXX35Bjx49MHfuXCQmJpr0fbt374758+cjMTERvr6+WL9+fb233zo4OGDOnDkIDw9HQEAA7O3tkZaWJr4/aNAgbN26FZmZmejduzf69u2L+fPno02bNvVe96GHHkJiYiJ69eqF3r1748SJE9i2bZvYpSGiB5dCMMeEMhEREdF94J8VREREJDsWJERERCQ7FiREREQkOxYkREREJDsWJERERCQ7FiREREQkOxYkREREJDsWJERERCQ7FiREREQkOxYkREREJDsWJERERCQ7FiREREQku/8Hb2l9kXeE0kUAAAAASUVORK5CYII=\n", |
||
| 2343 | "text/plain": [ |
||
| 2344 | "<Figure size 640x480 with 2 Axes>" |
||
| 2345 | ] |
||
| 2346 | }, |
||
| 2347 | "metadata": {}, |
||
| 2348 | "output_type": "display_data" |
||
| 2349 | } |
||
| 2350 | ], |
||
| 2351 | "source": [ |
||
| 2352 | "import seaborn as sns\n", |
||
| 2353 | "import matplotlib.pyplot as plt \n", |
||
| 2354 | "\n", |
||
| 2355 | "ax= plt.subplot()\n", |
||
| 2356 | "sns.heatmap(cm, annot=True, ax = ax); #annot=True to annotate cells\n", |
||
| 2357 | "\n", |
||
| 2358 | "# labels, title and ticks\n", |
||
| 2359 | "ax.set_xlabel('Predicted labels');ax.set_ylabel('True labels'); \n", |
||
| 2360 | "ax.set_title('Confusion Matrix'); " |
||
| 2361 | ] |
||
| 2362 | }, |
||
| 2363 | { |
||
| 2364 | "cell_type": "code", |
||
| 2365 | "execution_count": null, |
||
| 2366 | "metadata": {}, |
||
| 2367 | "outputs": [], |
||
| 2368 | "source": [] |
||
| 2369 | } |
||
| 2370 | ], |
||
| 2371 | "metadata": { |
||
| 2372 | "kernelspec": { |
||
| 2373 | "display_name": "Python 3 (ipykernel)", |
||
| 2374 | "language": "python", |
||
| 2375 | "name": "python3" |
||
| 2376 | }, |
||
| 2377 | "language_info": { |
||
| 2378 | "codemirror_mode": { |
||
| 2379 | "name": "ipython", |
||
| 2380 | "version": 3 |
||
| 2381 | }, |
||
| 2382 | "file_extension": ".py", |
||
| 2383 | "mimetype": "text/x-python", |
||
| 2384 | "name": "python", |
||
| 2385 | "nbconvert_exporter": "python", |
||
| 2386 | "pygments_lexer": "ipython3", |
||
| 2387 | "version": "3.10.9" |
||
| 2388 | } |
||
| 2389 | }, |
||
| 2390 | "nbformat": 4, |
||
| 2391 | "nbformat_minor": 4 |
||
| 2392 | } |