Hugging Face's logo

Wahb12111
/
Deepfake_Image_Classification

Image Classification
Keras
deepfake
computer-vision
efficientnet
Model card Files
xet
Community
Wahb12111 commited on
Commit
51e6e62
·
verified ·
1 Parent(s): 8d72ce5

Delete part1 (1).ipynb

Browse files
Files changed (1) hide show
  1. part1 (1).ipynb +0 -957
part1 (1).ipynb DELETED
@@ -1,957 +0,0 @@
1
- {
2
- "nbformat": 4,
3
- "nbformat_minor": 0,
4
- "metadata": {
5
- "colab": {
6
- "provenance": [],
7
- "gpuType": "T4"
8
- },
9
- "kernelspec": {
10
- "name": "python3",
11
- "display_name": "Python 3"
12
- },
13
- "language_info": {
14
- "name": "python"
15
- },
16
- "accelerator": "GPU"
17
- },
18
- "cells": [
19
- {
20
- "cell_type": "code",
21
- "execution_count": null,
22
- "metadata": {
23
- "colab": {
24
- "base_uri": "https://localhost:8080/"
25
- },
26
- "id": "_Q4gIvePGw85",
27
- "outputId": "21debd12-2d0d-43d4-9f0b-e79430f9a832"
28
- },
29
- "outputs": [
30
- {
31
- "output_type": "stream",
32
- "name": "stdout",
33
- "text": [
34
- "Mounted at /content/drive\n",
35
- "/content\n",
36
- " AssemblyTask\n",
37
- " BOOKS\n",
38
- " car2.jpg\n",
39
- " car.jpg\n",
40
- "'Colab Notebooks'\n",
41
- "'Copy of ⁨جدول الفرقة الثانية - عام - مج ٢.pdf⁩.pdf'\n",
42
- "'Coursera N24TOG3ZARUP.pdf'\n",
43
- " Grades\n",
44
- "'graphicsTask (2).png'\n",
45
- "'graphicsTask (2).svg'\n",
46
- " IMG_4899.jpeg\n",
47
- " IMG_4900.jpeg\n",
48
- " IMG_4901.jpeg\n",
49
- " IMG_5986.jpeg\n",
50
- " InternVideo.mp4\n",
51
- " LOOPS.gslides\n",
52
- " NTI-PRJCT\n",
53
- "'Session 6 .gdoc'\n",
54
- " SWEProject.drawio\n",
55
- " Task2.rar\n",
56
- "\"The _Animator's_Survival_Kit.pdf\"\n",
57
- "'Untitled Diagram'\n",
58
- "'Untitled document (1).gdoc'\n",
59
- "'Untitled document.gdoc'\n",
60
- " Wahb_CV.pdf\n",
61
- "'WhatsApp Image 2024-09-28 at 18.11.27_5e5ce6c9.jpg'\n",
62
- "'WhatsApp Image 2024-09-28 at 18.45.13_fd438e9f.jpg'\n",
63
- "'WhatsApp Image 2025-05-28 at 15.11 (1).24_84c876ae.jpg'\n",
64
- "'WhatsApp Image 2025-05-28 at 15.11.24_84c876ae.jpg'\n"
65
- ]
66
- }
67
- ],
68
- "source": [
69
- "from google.colab import drive\n",
70
- "drive.mount('/content/drive')\n",
71
- "\n",
72
- "import os\n",
73
- "print(os.getcwd())\n",
74
- "\n",
75
- "!ls /content/drive/MyDrive/"
76
- ]
77
- },
78
- {
79
- "cell_type": "code",
80
- "source": [
81
- "from google.colab import files\n",
82
- "files.upload()\n",
83
- "\n",
84
- "!pip install kaggle\n",
85
- "\n",
86
- "!mkdir -p ~/.kaggle\n",
87
- "!cp kaggle.json ~/.kaggle/\n",
88
- "!chmod 600 ~/.kaggle/kaggle.json\n",
89
- "\n",
90
- "!kaggle datasets download -d dagnelies/deepfake-faces\n",
91
- "!unzip -q deepfake-faces.zip -d deepfake_faces\n"
92
- ],
93
- "metadata": {
94
- "colab": {
95
- "base_uri": "https://localhost:8080/",
96
- "height": 440
97
- },
98
- "id": "7Emqe4aNG7Ak",
99
- "outputId": "769def2b-ae6c-491e-8cf3-57c4b1e678d4"
100
- },
101
- "execution_count": null,
102
- "outputs": [
103
- {
104
- "output_type": "display_data",
105
- "data": {
106
- "text/plain": [
107
- "<IPython.core.display.HTML object>"
108
- ],
109
- "text/html": [
110
- "\n",
111
- " <input type=\"file\" id=\"files-f634150d-7f42-49fc-8650-c73cced50986\" name=\"files[]\" multiple disabled\n",
112
- " style=\"border:none\" />\n",
113
- " <output id=\"result-f634150d-7f42-49fc-8650-c73cced50986\">\n",
114
- " Upload widget is only available when the cell has been executed in the\n",
115
- " current browser session. Please rerun this cell to enable.\n",
116
- " </output>\n",
117
- " <script>// Copyright 2017 Google LLC\n",
118
- "//\n",
119
- "// Licensed under the Apache License, Version 2.0 (the \"License\");\n",
120
- "// you may not use this file except in compliance with the License.\n",
121
- "// You may obtain a copy of the License at\n",
122
- "//\n",
123
- "// http://www.apache.org/licenses/LICENSE-2.0\n",
124
- "//\n",
125
- "// Unless required by applicable law or agreed to in writing, software\n",
126
- "// distributed under the License is distributed on an \"AS IS\" BASIS,\n",
127
- "// WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.\n",
128
- "// See the License for the specific language governing permissions and\n",
129
- "// limitations under the License.\n",
130
- "\n",
131
- "/**\n",
132
- " * @fileoverview Helpers for google.colab Python module.\n",
133
- " */\n",
134
- "(function(scope) {\n",
135
- "function span(text, styleAttributes = {}) {\n",
136
- " const element = document.createElement('span');\n",
137
- " element.textContent = text;\n",
138
- " for (const key of Object.keys(styleAttributes)) {\n",
139
- " element.style[key] = styleAttributes[key];\n",
140
- " }\n",
141
- " return element;\n",
142
- "}\n",
143
- "\n",
144
- "// Max number of bytes which will be uploaded at a time.\n",
145
- "const MAX_PAYLOAD_SIZE = 100 * 1024;\n",
146
- "\n",
147
- "function _uploadFiles(inputId, outputId) {\n",
148
- " const steps = uploadFilesStep(inputId, outputId);\n",
149
- " const outputElement = document.getElementById(outputId);\n",
150
- " // Cache steps on the outputElement to make it available for the next call\n",
151
- " // to uploadFilesContinue from Python.\n",
152
- " outputElement.steps = steps;\n",
153
- "\n",
154
- " return _uploadFilesContinue(outputId);\n",
155
- "}\n",
156
- "\n",
157
- "// This is roughly an async generator (not supported in the browser yet),\n",
158
- "// where there are multiple asynchronous steps and the Python side is going\n",
159
- "// to poll for completion of each step.\n",
160
- "// This uses a Promise to block the python side on completion of each step,\n",
161
- "// then passes the result of the previous step as the input to the next step.\n",
162
- "function _uploadFilesContinue(outputId) {\n",
163
- " const outputElement = document.getElementById(outputId);\n",
164
- " const steps = outputElement.steps;\n",
165
- "\n",
166
- " const next = steps.next(outputElement.lastPromiseValue);\n",
167
- " return Promise.resolve(next.value.promise).then((value) => {\n",
168
- " // Cache the last promise value to make it available to the next\n",
169
- " // step of the generator.\n",
170
- " outputElement.lastPromiseValue = value;\n",
171
- " return next.value.response;\n",
172
- " });\n",
173
- "}\n",
174
- "\n",
175
- "/**\n",
176
- " * Generator function which is called between each async step of the upload\n",
177
- " * process.\n",
178
- " * @param {string} inputId Element ID of the input file picker element.\n",
179
- " * @param {string} outputId Element ID of the output display.\n",
180
- " * @return {!Iterable<!Object>} Iterable of next steps.\n",
181
- " */\n",
182
- "function* uploadFilesStep(inputId, outputId) {\n",
183
- " const inputElement = document.getElementById(inputId);\n",
184
- " inputElement.disabled = false;\n",
185
- "\n",
186
- " const outputElement = document.getElementById(outputId);\n",
187
- " outputElement.innerHTML = '';\n",
188
- "\n",
189
- " const pickedPromise = new Promise((resolve) => {\n",
190
- " inputElement.addEventListener('change', (e) => {\n",
191
- " resolve(e.target.files);\n",
192
- " });\n",
193
- " });\n",
194
- "\n",
195
- " const cancel = document.createElement('button');\n",
196
- " inputElement.parentElement.appendChild(cancel);\n",
197
- " cancel.textContent = 'Cancel upload';\n",
198
- " const cancelPromise = new Promise((resolve) => {\n",
199
- " cancel.onclick = () => {\n",
200
- " resolve(null);\n",
201
- " };\n",
202
- " });\n",
203
- "\n",
204
- " // Wait for the user to pick the files.\n",
205
- " const files = yield {\n",
206
- " promise: Promise.race([pickedPromise, cancelPromise]),\n",
207
- " response: {\n",
208
- " action: 'starting',\n",
209
- " }\n",
210
- " };\n",
211
- "\n",
212
- " cancel.remove();\n",
213
- "\n",
214
- " // Disable the input element since further picks are not allowed.\n",
215
- " inputElement.disabled = true;\n",
216
- "\n",
217
- " if (!files) {\n",
218
- " return {\n",
219
- " response: {\n",
220
- " action: 'complete',\n",
221
- " }\n",
222
- " };\n",
223
- " }\n",
224
- "\n",
225
- " for (const file of files) {\n",
226
- " const li = document.createElement('li');\n",
227
- " li.append(span(file.name, {fontWeight: 'bold'}));\n",
228
- " li.append(span(\n",
229
- " `(${file.type || 'n/a'}) - ${file.size} bytes, ` +\n",
230
- " `last modified: ${\n",
231
- " file.lastModifiedDate ? file.lastModifiedDate.toLocaleDateString() :\n",
232
- " 'n/a'} - `));\n",
233
- " const percent = span('0% done');\n",
234
- " li.appendChild(percent);\n",
235
- "\n",
236
- " outputElement.appendChild(li);\n",
237
- "\n",
238
- " const fileDataPromise = new Promise((resolve) => {\n",
239
- " const reader = new FileReader();\n",
240
- " reader.onload = (e) => {\n",
241
- " resolve(e.target.result);\n",
242
- " };\n",
243
- " reader.readAsArrayBuffer(file);\n",
244
- " });\n",
245
- " // Wait for the data to be ready.\n",
246
- " let fileData = yield {\n",
247
- " promise: fileDataPromise,\n",
248
- " response: {\n",
249
- " action: 'continue',\n",
250
- " }\n",
251
- " };\n",
252
- "\n",
253
- " // Use a chunked sending to avoid message size limits. See b/62115660.\n",
254
- " let position = 0;\n",
255
- " do {\n",
256
- " const length = Math.min(fileData.byteLength - position, MAX_PAYLOAD_SIZE);\n",
257
- " const chunk = new Uint8Array(fileData, position, length);\n",
258
- " position += length;\n",
259
- "\n",
260
- " const base64 = btoa(String.fromCharCode.apply(null, chunk));\n",
261
- " yield {\n",
262
- " response: {\n",
263
- " action: 'append',\n",
264
- " file: file.name,\n",
265
- " data: base64,\n",
266
- " },\n",
267
- " };\n",
268
- "\n",
269
- " let percentDone = fileData.byteLength === 0 ?\n",
270
- " 100 :\n",
271
- " Math.round((position / fileData.byteLength) * 100);\n",
272
- " percent.textContent = `${percentDone}% done`;\n",
273
- "\n",
274
- " } while (position < fileData.byteLength);\n",
275
- " }\n",
276
- "\n",
277
- " // All done.\n",
278
- " yield {\n",
279
- " response: {\n",
280
- " action: 'complete',\n",
281
- " }\n",
282
- " };\n",
283
- "}\n",
284
- "\n",
285
- "scope.google = scope.google || {};\n",
286
- "scope.google.colab = scope.google.colab || {};\n",
287
- "scope.google.colab._files = {\n",
288
- " _uploadFiles,\n",
289
- " _uploadFilesContinue,\n",
290
- "};\n",
291
- "})(self);\n",
292
- "</script> "
293
- ]
294
- },
295
- "metadata": {}
296
- },
297
- {
298
- "output_type": "stream",
299
- "name": "stdout",
300
- "text": [
301
- "Saving kaggle.json to kaggle.json\n",
302
- "Requirement already satisfied: kaggle in /usr/local/lib/python3.11/dist-packages (1.7.4.5)\n",
303
- "Requirement already satisfied: bleach in /usr/local/lib/python3.11/dist-packages (from kaggle) (6.2.0)\n",
304
- "Requirement already satisfied: certifi>=14.05.14 in /usr/local/lib/python3.11/dist-packages (from kaggle) (2025.8.3)\n",
305
- "Requirement already satisfied: charset-normalizer in /usr/local/lib/python3.11/dist-packages (from kaggle) (3.4.3)\n",
306
- "Requirement already satisfied: idna in /usr/local/lib/python3.11/dist-packages (from kaggle) (3.10)\n",
307
- "Requirement already satisfied: protobuf in /usr/local/lib/python3.11/dist-packages (from kaggle) (5.29.5)\n",
308
- "Requirement already satisfied: python-dateutil>=2.5.3 in /usr/local/lib/python3.11/dist-packages (from kaggle) (2.9.0.post0)\n",
309
- "Requirement already satisfied: python-slugify in /usr/local/lib/python3.11/dist-packages (from kaggle) (8.0.4)\n",
310
- "Requirement already satisfied: requests in /usr/local/lib/python3.11/dist-packages (from kaggle) (2.32.3)\n",
311
- "Requirement already satisfied: setuptools>=21.0.0 in /usr/local/lib/python3.11/dist-packages (from kaggle) (75.2.0)\n",
312
- "Requirement already satisfied: six>=1.10 in /usr/local/lib/python3.11/dist-packages (from kaggle) (1.17.0)\n",
313
- "Requirement already satisfied: text-unidecode in /usr/local/lib/python3.11/dist-packages (from kaggle) (1.3)\n",
314
- "Requirement already satisfied: tqdm in /usr/local/lib/python3.11/dist-packages (from kaggle) (4.67.1)\n",
315
- "Requirement already satisfied: urllib3>=1.15.1 in /usr/local/lib/python3.11/dist-packages (from kaggle) (2.5.0)\n",
316
- "Requirement already satisfied: webencodings in /usr/local/lib/python3.11/dist-packages (from kaggle) (0.5.1)\n",
317
- "Dataset URL: https://www.kaggle.com/datasets/dagnelies/deepfake-faces\n",
318
- "License(s): other\n",
319
- "Downloading deepfake-faces.zip to /content\n",
320
- " 96% 416M/433M [00:00<00:00, 447MB/s]\n",
321
- "100% 433M/433M [00:00<00:00, 516MB/s]\n"
322
- ]
323
- }
324
- ]
325
- },
326
- {
327
- "cell_type": "code",
328
- "source": [
329
- "import numpy as np\n",
330
- "import matplotlib.pyplot as plt\n",
331
- "import pandas as pd\n",
332
- "import seaborn as sns\n",
333
- "import plotly.graph_objects as go\n",
334
- "from plotly.offline import iplot\n",
335
- "import tensorflow as tf\n",
336
- "from sklearn.model_selection import train_test_split\n",
337
- "import cv2"
338
- ],
339
- "metadata": {
340
- "id": "sgQ6ibXHHCAd"
341
- },
342
- "execution_count": null,
343
- "outputs": []
344
- },
345
- {
346
- "cell_type": "code",
347
- "source": [
348
- "meta = pd.read_csv('/content/deepfake_faces/metadata.csv')\n",
349
- "print(meta.head())"
350
- ],
351
- "metadata": {
352
- "colab": {
353
- "base_uri": "https://localhost:8080/"
354
- },
355
- "id": "HRchwl81HEiU",
356
- "outputId": "efb7634a-6f85-4be3-ed9e-be16005c75b4"
357
- },
358
- "execution_count": null,
359
- "outputs": [
360
- {
361
- "output_type": "stream",
362
- "name": "stdout",
363
- "text": [
364
- " videoname original_width original_height label original\n",
365
- "0 aznyksihgl.mp4 129 129 FAKE xnojggkrxt.mp4\n",
366
- "1 gkwmalrvcj.mp4 129 129 FAKE hqqmtxvbjj.mp4\n",
367
- "2 lxnqzocgaq.mp4 223 217 FAKE xjzkfqddyk.mp4\n",
368
- "3 itsbtrrelv.mp4 186 186 FAKE kqvepwqxfe.mp4\n",
369
- "4 ddvgrczjno.mp4 155 155 FAKE pluadmqqta.mp4\n"
370
- ]
371
- }
372
- ]
373
- },
374
- {
375
- "cell_type": "code",
376
- "source": [
377
- "def summary(df):\n",
378
- " summary_df = pd.DataFrame(df.dtypes, columns=['dtypes'])\n",
379
- " summary_df['count'] = df.count().values\n",
380
- " summary_df['unique'] = df.nunique().values\n",
381
- " summary_df['missing#'] = df.isna().sum()\n",
382
- " summary_df['missing%'] = df.isna().sum() / len(df)\n",
383
- " return summary_df"
384
- ],
385
- "metadata": {
386
- "id": "8da4BT5CHGVz"
387
- },
388
- "execution_count": null,
389
- "outputs": []
390
- },
391
- {
392
- "cell_type": "code",
393
- "source": [
394
- "print(summary(meta).style.background_gradient('Purples'))\n",
395
- "\n",
396
- "print('Fake Images:', len(meta[meta.label=='FAKE']))\n",
397
- "print('Real Images:', len(meta[meta.label=='REAL']))\n"
398
- ],
399
- "metadata": {
400
- "colab": {
401
- "base_uri": "https://localhost:8080/"
402
- },
403
- "id": "SaZKTtqFHKQM",
404
- "outputId": "088c301f-1b19-40c9-f210-d09baea550d8"
405
- },
406
- "execution_count": null,
407
- "outputs": [
408
- {
409
- "output_type": "stream",
410
- "name": "stdout",
411
- "text": [
412
- "<pandas.io.formats.style.Styler object at 0x7dae5a9e74d0>\n",
413
- "Fake Images: 79341\n",
414
- "Real Images: 16293\n"
415
- ]
416
- }
417
- ]
418
- },
419
- {
420
- "cell_type": "code",
421
- "source": [
422
- "# Sample balanced dataset\n",
423
- "real_df = meta[meta['label'] == 'REAL']\n",
424
- "fake_df = meta[meta['label'] == 'FAKE']\n",
425
- "sample_size = 16000\n",
426
- "\n",
427
- "real_df = real_df.sample(sample_size, random_state=42)\n",
428
- "fake_df = fake_df.sample(sample_size, random_state=42)\n",
429
- "\n",
430
- "sample_meta = pd.concat([real_df, fake_df])"
431
- ],
432
- "metadata": {
433
- "id": "9m8dhK0ZHP8c"
434
- },
435
- "execution_count": null,
436
- "outputs": []
437
- },
438
- {
439
- "cell_type": "code",
440
- "source": [
441
- "sample_meta['filepath'] = '/content/deepfake_faces/faces_224/' + sample_meta['videoname'].str[:-4] + '.jpg'\n",
442
- "\n",
443
- "# Split the data\n",
444
- "Train_set, temp_df = train_test_split(sample_meta, test_size=0.2, random_state=42, stratify=sample_meta['label'])\n",
445
- "Val_set, Test_set = train_test_split(temp_df, test_size=0.5, random_state=42, stratify=temp_df['label'])\n",
446
- "\n",
447
- "print(f'Train Set: {Train_set.shape}')\n",
448
- "print(f'Validation Set: {Val_set.shape}')\n",
449
- "print(f'Test Set: {Test_set.shape}')"
450
- ],
451
- "metadata": {
452
- "colab": {
453
- "base_uri": "https://localhost:8080/"
454
- },
455
- "id": "bL62yOEcHWqE",
456
- "outputId": "1a8042a4-0115-4821-b4b7-077b1bcbaf35"
457
- },
458
- "execution_count": null,
459
- "outputs": [
460
- {
461
- "output_type": "stream",
462
- "name": "stdout",
463
- "text": [
464
- "Train Set: (25600, 6)\n",
465
- "Validation Set: (3200, 6)\n",
466
- "Test Set: (3200, 6)\n"
467
- ]
468
- }
469
- ]
470
- },
471
- {
472
- "cell_type": "code",
473
- "source": [
474
- "data_augmentation = tf.keras.Sequential([\n",
475
- " tf.keras.layers.RandomFlip(\"horizontal\"),\n",
476
- " tf.keras.layers.RandomRotation(0.1),\n",
477
- " tf.keras.layers.RandomZoom(0.1),\n",
478
- "])"
479
- ],
480
- "metadata": {
481
- "id": "e4xmsSeGIDe8"
482
- },
483
- "execution_count": null,
484
- "outputs": []
485
- },
486
- {
487
- "cell_type": "code",
488
- "source": [
489
- "label_map = {'REAL': 0, 'FAKE': 1}\n"
490
- ],
491
- "metadata": {
492
- "id": "2ba3DFLDIFWs"
493
- },
494
- "execution_count": null,
495
- "outputs": []
496
- },
497
- {
498
- "cell_type": "code",
499
- "source": [
500
- "def preprocess_image(filepath, label):\n",
501
- " image = tf.io.read_file(filepath)\n",
502
- " image = tf.image.decode_jpeg(image, channels=3)\n",
503
- " image = tf.image.resize(image, [224, 224])\n",
504
- " image = tf.cast(image, tf.float32)\n",
505
- " # EfficientNet expects values in [0, 255] for its preprocess function\n",
506
- " return image, label\n",
507
- "\n",
508
- "# Create TensorFlow datasets\n",
509
- "batch_size = 32\n",
510
- "\n",
511
- "train_ds = tf.data.Dataset.from_tensor_slices((\n",
512
- " Train_set['filepath'].values,\n",
513
- " Train_set['label'].map(label_map).values\n",
514
- "))\n",
515
- "train_ds = (train_ds\n",
516
- " .map(preprocess_image, num_parallel_calls=tf.data.AUTOTUNE)\n",
517
- " .shuffle(1000, seed=42)\n",
518
- " .batch(batch_size)\n",
519
- " .prefetch(tf.data.AUTOTUNE))\n",
520
- "\n",
521
- "val_ds = tf.data.Dataset.from_tensor_slices((\n",
522
- " Val_set['filepath'].values,\n",
523
- " Val_set['label'].map(label_map).values\n",
524
- "))\n",
525
- "val_ds = (val_ds\n",
526
- " .map(preprocess_image, num_parallel_calls=tf.data.AUTOTUNE)\n",
527
- " .batch(batch_size)\n",
528
- " .prefetch(tf.data.AUTOTUNE))\n",
529
- "\n",
530
- "test_ds = tf.data.Dataset.from_tensor_slices((\n",
531
- " Test_set['filepath'].values,\n",
532
- " Test_set['label'].map(label_map).values\n",
533
- "))\n",
534
- "test_ds = (test_ds\n",
535
- " .map(preprocess_image, num_parallel_calls=tf.data.AUTOTUNE)\n",
536
- " .batch(batch_size)\n",
537
- " .prefetch(tf.data.AUTOTUNE))"
538
- ],
539
- "metadata": {
540
- "id": "OendSPcjIF99"
541
- },
542
- "execution_count": null,
543
- "outputs": []
544
- },
545
- {
546
- "cell_type": "code",
547
- "source": [
548
- "def plot_class_counts(train_set, val_set, test_set):\n",
549
- " sets = ['Train Set', 'Validation Set', 'Test Set']\n",
550
- " colors = ['#52A666', '#C15B4E']\n",
551
- "\n",
552
- " y = {\n",
553
- " 'REAL': [np.sum(train_set == 'REAL'), np.sum(val_set == 'REAL'), np.sum(test_set == 'REAL')],\n",
554
- " 'FAKE': [np.sum(train_set == 'FAKE'), np.sum(val_set == 'FAKE'), np.sum(test_set == 'FAKE')]\n",
555
- " }\n",
556
- "\n",
557
- " trace0 = go.Bar(x=sets, y=y['REAL'], name='REAL', marker={'color': colors[0]}, opacity=0.7)\n",
558
- " trace1 = go.Bar(x=sets, y=y['FAKE'], name='FAKE', marker={'color': colors[1]}, opacity=0.7)\n",
559
- "\n",
560
- " data = [trace0, trace1]\n",
561
- " layout = go.Layout(title='Count of Classes in each set:', xaxis={'title': 'Set'}, yaxis={'title': 'Count'})\n",
562
- "\n",
563
- " fig = go.Figure(data, layout)\n",
564
- " iplot(fig)\n",
565
- "\n",
566
- "plot_class_counts(np.array(Train_set['label']), np.array(Val_set['label']), np.array(Test_set['label']))\n"
567
- ],
568
- "metadata": {
569
- "colab": {
570
- "base_uri": "https://localhost:8080/",
571
- "height": 542
572
- },
573
- "id": "BUtvuXUPIL3E",
574
- "outputId": "a7261012-adf3-4212-fac0-1b37f0fd8159"
575
- },
576
- "execution_count": null,
577
- "outputs": [
578
- {
579
- "output_type": "display_data",
580
- "data": {
581
- "text/html": [
582
- "<html>\n",
583
- "<head><meta charset=\"utf-8\" /></head>\n",
584
- "<body>\n",
585
- " <div> <script src=\"https://cdnjs.cloudflare.com/ajax/libs/mathjax/2.7.5/MathJax.js?config=TeX-AMS-MML_SVG\"></script><script type=\"text/javascript\">if (window.MathJax && window.MathJax.Hub && window.MathJax.Hub.Config) {window.MathJax.Hub.Config({SVG: {font: \"STIX-Web\"}});}</script> <script type=\"text/javascript\">window.PlotlyConfig = {MathJaxConfig: 'local'};</script>\n",
586
- " <script charset=\"utf-8\" src=\"https://cdn.plot.ly/plotly-2.35.2.min.js\"></script> <div id=\"3aa0cb85-3812-44b8-a6a0-293506fec333\" class=\"plotly-graph-div\" style=\"height:525px; width:100%;\"></div> <script type=\"text/javascript\"> window.PLOTLYENV=window.PLOTLYENV || {}; if (document.getElementById(\"3aa0cb85-3812-44b8-a6a0-293506fec333\")) { Plotly.newPlot( \"3aa0cb85-3812-44b8-a6a0-293506fec333\", [{\"marker\":{\"color\":\"#52A666\"},\"name\":\"REAL\",\"opacity\":0.7,\"x\":[\"Train Set\",\"Validation Set\",\"Test Set\"],\"y\":[12800,1600,1600],\"type\":\"bar\"},{\"marker\":{\"color\":\"#C15B4E\"},\"name\":\"FAKE\",\"opacity\":0.7,\"x\":[\"Train Set\",\"Validation Set\",\"Test Set\"],\"y\":[12800,1600,1600],\"type\":\"bar\"}], {\"template\":{\"data\":{\"barpolar\":[{\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"barpolar\"}],\"bar\":[{\"error_x\":{\"color\":\"#2a3f5f\"},\"error_y\":{\"color\":\"#2a3f5f\"},\"marker\":{\"line\":{\"color\":\"#E5ECF6\",\"width\":0.5},\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"bar\"}],\"carpet\":[{\"aaxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"baxis\":{\"endlinecolor\":\"#2a3f5f\",\"gridcolor\":\"white\",\"linecolor\":\"white\",\"minorgridcolor\":\"white\",\"startlinecolor\":\"#2a3f5f\"},\"type\":\"carpet\"}],\"choropleth\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"choropleth\"}],\"contourcarpet\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"contourcarpet\"}],\"contour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"contour\"}],\"heatmapgl\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmapgl\"}],\"heatmap\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"heatmap\"}],\"histogram2dcontour\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2dcontour\"}],\"histogram2d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"histogram2d\"}],\"histogram\":[{\"marker\":{\"pattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2}},\"type\":\"histogram\"}],\"mesh3d\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"type\":\"mesh3d\"}],\"parcoords\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"parcoords\"}],\"pie\":[{\"automargin\":true,\"type\":\"pie\"}],\"scatter3d\":[{\"line\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatter3d\"}],\"scattercarpet\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattercarpet\"}],\"scattergeo\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergeo\"}],\"scattergl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattergl\"}],\"scattermapbox\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scattermapbox\"}],\"scatterpolargl\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolargl\"}],\"scatterpolar\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterpolar\"}],\"scatter\":[{\"fillpattern\":{\"fillmode\":\"overlay\",\"size\":10,\"solidity\":0.2},\"type\":\"scatter\"}],\"scatterternary\":[{\"marker\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"type\":\"scatterternary\"}],\"surface\":[{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"},\"colorscale\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"type\":\"surface\"}],\"table\":[{\"cells\":{\"fill\":{\"color\":\"#EBF0F8\"},\"line\":{\"color\":\"white\"}},\"header\":{\"fill\":{\"color\":\"#C8D4E3\"},\"line\":{\"color\":\"white\"}},\"type\":\"table\"}]},\"layout\":{\"annotationdefaults\":{\"arrowcolor\":\"#2a3f5f\",\"arrowhead\":0,\"arrowwidth\":1},\"autotypenumbers\":\"strict\",\"coloraxis\":{\"colorbar\":{\"outlinewidth\":0,\"ticks\":\"\"}},\"colorscale\":{\"diverging\":[[0,\"#8e0152\"],[0.1,\"#c51b7d\"],[0.2,\"#de77ae\"],[0.3,\"#f1b6da\"],[0.4,\"#fde0ef\"],[0.5,\"#f7f7f7\"],[0.6,\"#e6f5d0\"],[0.7,\"#b8e186\"],[0.8,\"#7fbc41\"],[0.9,\"#4d9221\"],[1,\"#276419\"]],\"sequential\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]],\"sequentialminus\":[[0.0,\"#0d0887\"],[0.1111111111111111,\"#46039f\"],[0.2222222222222222,\"#7201a8\"],[0.3333333333333333,\"#9c179e\"],[0.4444444444444444,\"#bd3786\"],[0.5555555555555556,\"#d8576b\"],[0.6666666666666666,\"#ed7953\"],[0.7777777777777778,\"#fb9f3a\"],[0.8888888888888888,\"#fdca26\"],[1.0,\"#f0f921\"]]},\"colorway\":[\"#636efa\",\"#EF553B\",\"#00cc96\",\"#ab63fa\",\"#FFA15A\",\"#19d3f3\",\"#FF6692\",\"#B6E880\",\"#FF97FF\",\"#FECB52\"],\"font\":{\"color\":\"#2a3f5f\"},\"geo\":{\"bgcolor\":\"white\",\"lakecolor\":\"white\",\"landcolor\":\"#E5ECF6\",\"showlakes\":true,\"showland\":true,\"subunitcolor\":\"white\"},\"hoverlabel\":{\"align\":\"left\"},\"hovermode\":\"closest\",\"mapbox\":{\"style\":\"light\"},\"paper_bgcolor\":\"white\",\"plot_bgcolor\":\"#E5ECF6\",\"polar\":{\"angularaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"radialaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"scene\":{\"xaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"yaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"},\"zaxis\":{\"backgroundcolor\":\"#E5ECF6\",\"gridcolor\":\"white\",\"gridwidth\":2,\"linecolor\":\"white\",\"showbackground\":true,\"ticks\":\"\",\"zerolinecolor\":\"white\"}},\"shapedefaults\":{\"line\":{\"color\":\"#2a3f5f\"}},\"ternary\":{\"aaxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"baxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"},\"bgcolor\":\"#E5ECF6\",\"caxis\":{\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\"}},\"title\":{\"x\":0.05},\"xaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2},\"yaxis\":{\"automargin\":true,\"gridcolor\":\"white\",\"linecolor\":\"white\",\"ticks\":\"\",\"title\":{\"standoff\":15},\"zerolinecolor\":\"white\",\"zerolinewidth\":2}}},\"title\":{\"text\":\"Count of Classes in each set:\"},\"xaxis\":{\"title\":{\"text\":\"Set\"}},\"yaxis\":{\"title\":{\"text\":\"Count\"}}}, {\"responsive\": true} ).then(function(){\n",
587
- " \n",
588
- "var gd = document.getElementById('3aa0cb85-3812-44b8-a6a0-293506fec333');\n",
589
- "var x = new MutationObserver(function (mutations, observer) {{\n",
590
- " var display = window.getComputedStyle(gd).display;\n",
591
- " if (!display || display === 'none') {{\n",
592
- " console.log([gd, 'removed!']);\n",
593
- " Plotly.purge(gd);\n",
594
- " observer.disconnect();\n",
595
- " }}\n",
596
- "}});\n",
597
- "\n",
598
- "// Listen for the removal of the full notebook cells\n",
599
- "var notebookContainer = gd.closest('#notebook-container');\n",
600
- "if (notebookContainer) {{\n",
601
- " x.observe(notebookContainer, {childList: true});\n",
602
- "}}\n",
603
- "\n",
604
- "// Listen for the clearing of the current output cell\n",
605
- "var outputEl = gd.closest('.output');\n",
606
- "if (outputEl) {{\n",
607
- " x.observe(outputEl, {childList: true});\n",
608
- "}}\n",
609
- "\n",
610
- " }) }; </script> </div>\n",
611
- "</body>\n",
612
- "</html>"
613
- ]
614
- },
615
- "metadata": {}
616
- }
617
- ]
618
- },
619
- {
620
- "cell_type": "code",
621
- "source": [
622
- "tf.keras.backend.clear_session()\n",
623
- "tf.random.set_seed(42)\n",
624
- "\n",
625
- "from tensorflow.keras.applications import EfficientNetB4\n",
626
- "\n",
627
- "# Build model with data augmentation\n",
628
- "base_model = EfficientNetB4(include_top=False, weights='imagenet', input_shape=(224, 224, 3))\n",
629
- "\n",
630
- "# Create the full model with data augmentation\n",
631
- "inputs = tf.keras.Input(shape=(224, 224, 3))\n",
632
- "\n",
633
- "# Apply data augmentation only during training\n",
634
- "x = data_augmentation(inputs)\n",
635
- "\n",
636
- "# Apply EfficientNet preprocessing\n",
637
- "x = tf.keras.applications.efficientnet.preprocess_input(x)\n",
638
- "\n",
639
- "# Pass through base model\n",
640
- "x = base_model(x, training=False) # Keep base model frozen initially\n",
641
- "\n",
642
- "# Add classification head\n",
643
- "x = tf.keras.layers.GlobalAveragePooling2D()(x)\n",
644
- "x = tf.keras.layers.Dropout(0.2)(x)\n",
645
- "outputs = tf.keras.layers.Dense(1, activation=\"sigmoid\")(x)\n",
646
- "\n",
647
- "model = tf.keras.Model(inputs, outputs)"
648
- ],
649
- "metadata": {
650
- "colab": {
651
- "base_uri": "https://localhost:8080/"
652
- },
653
- "id": "ev8RU-JIIU4M",
654
- "outputId": "7cbc8dd3-3ad7-462b-ff76-fd5f1909d11e"
655
- },
656
- "execution_count": null,
657
- "outputs": [
658
- {
659
- "output_type": "stream",
660
- "name": "stdout",
661
- "text": [
662
- "Downloading data from https://storage.googleapis.com/keras-applications/efficientnetb4_notop.h5\n",
663
- "\u001b[1m71686520/71686520\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m1s\u001b[0m 0us/step\n"
664
- ]
665
- }
666
- ]
667
- },
668
- {
669
- "cell_type": "code",
670
- "source": [
671
- "# Compile model\n",
672
- "optimizer = tf.keras.optimizers.Adam(learning_rate=0.001) # Using Adam for better convergence\n",
673
- "model.compile(\n",
674
- " optimizer=optimizer,\n",
675
- " loss=\"binary_crossentropy\",\n",
676
- " metrics=[\"accuracy\"]\n",
677
- ")\n",
678
- "\n",
679
- "model.summary()"
680
- ],
681
- "metadata": {
682
- "colab": {
683
- "base_uri": "https://localhost:8080/",
684
- "height": 357
685
- },
686
- "id": "0IpweVPuIY38",
687
- "outputId": "971546e4-c2c5-4d3e-d4c6-b618b5e3e58e"
688
- },
689
- "execution_count": null,
690
- "outputs": [
691
- {
692
- "output_type": "display_data",
693
- "data": {
694
- "text/plain": [
695
- "\u001b[1mModel: \"functional_1\"\u001b[0m\n"
696
- ],
697
- "text/html": [
698
- "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\">Model: \"functional_1\"</span>\n",
699
- "</pre>\n"
700
- ]
701
- },
702
- "metadata": {}
703
- },
704
- {
705
- "output_type": "display_data",
706
- "data": {
707
- "text/plain": [
708
- "┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
709
- "┃\u001b[1m \u001b[0m\u001b[1mLayer (type) \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1mOutput Shape \u001b[0m\u001b[1m \u001b[0m┃\u001b[1m \u001b[0m\u001b[1m Param #\u001b[0m\u001b[1m \u001b[0m┃\n",
710
- "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
711
- "│ input_layer_1 (\u001b[38;5;33mInputLayer\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m3\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
712
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
713
- "│ sequential (\u001b[38;5;33mSequential\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m224\u001b[0m, \u001b[38;5;34m3\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
714
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
715
- "│ efficientnetb4 (\u001b[38;5;33mFunctional\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m7\u001b[0m, \u001b[38;5;34m1792\u001b[0m) │ \u001b[38;5;34m17,673,823\u001b[0m │\n",
716
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
717
- "│ global_average_pooling2d │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1792\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
718
- "│ (\u001b[38;5;33mGlobalAveragePooling2D\u001b[0m) │ │ │\n",
719
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
720
- "│ dropout (\u001b[38;5;33mDropout\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1792\u001b[0m) │ \u001b[38;5;34m0\u001b[0m │\n",
721
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
722
- "│ dense (\u001b[38;5;33mDense\u001b[0m) │ (\u001b[38;5;45mNone\u001b[0m, \u001b[38;5;34m1\u001b[0m) │ \u001b[38;5;34m1,793\u001b[0m │\n",
723
- "└─────────────────────────────────┴────────────────────────┴───────────────┘\n"
724
- ],
725
- "text/html": [
726
- "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">┏━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━┓\n",
727
- "┃<span style=\"font-weight: bold\"> Layer (type) </span>┃<span style=\"font-weight: bold\"> Output Shape </span>┃<span style=\"font-weight: bold\"> Param # </span>┃\n",
728
- "┡━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━┩\n",
729
- "│ input_layer_1 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">InputLayer</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
730
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
731
- "│ sequential (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Sequential</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">224</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">3</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
732
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
733
- "│ efficientnetb4 (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Functional</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">7</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1792</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">17,673,823</span> │\n",
734
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
735
- "│ global_average_pooling2d │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1792</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
736
- "│ (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">GlobalAveragePooling2D</span>) │ │ │\n",
737
- "├─────────────────────────────────┼────────────────────────┼──────────────���┤\n",
738
- "│ dropout (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dropout</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1792</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">0</span> │\n",
739
- "├─────────────────────────────────┼────────────────────────┼───────────────┤\n",
740
- "│ dense (<span style=\"color: #0087ff; text-decoration-color: #0087ff\">Dense</span>) │ (<span style=\"color: #00d7ff; text-decoration-color: #00d7ff\">None</span>, <span style=\"color: #00af00; text-decoration-color: #00af00\">1</span>) │ <span style=\"color: #00af00; text-decoration-color: #00af00\">1,793</span> │\n",
741
- "└─────────────────────────────────┴────────────────────────┴───────────────┘\n",
742
- "</pre>\n"
743
- ]
744
- },
745
- "metadata": {}
746
- },
747
- {
748
- "output_type": "display_data",
749
- "data": {
750
- "text/plain": [
751
- "\u001b[1m Total params: \u001b[0m\u001b[38;5;34m17,675,616\u001b[0m (67.43 MB)\n"
752
- ],
753
- "text/html": [
754
- "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Total params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">17,675,616</span> (67.43 MB)\n",
755
- "</pre>\n"
756
- ]
757
- },
758
- "metadata": {}
759
- },
760
- {
761
- "output_type": "display_data",
762
- "data": {
763
- "text/plain": [
764
- "\u001b[1m Trainable params: \u001b[0m\u001b[38;5;34m17,550,409\u001b[0m (66.95 MB)\n"
765
- ],
766
- "text/html": [
767
- "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">17,550,409</span> (66.95 MB)\n",
768
- "</pre>\n"
769
- ]
770
- },
771
- "metadata": {}
772
- },
773
- {
774
- "output_type": "display_data",
775
- "data": {
776
- "text/plain": [
777
- "\u001b[1m Non-trainable params: \u001b[0m\u001b[38;5;34m125,207\u001b[0m (489.09 KB)\n"
778
- ],
779
- "text/html": [
780
- "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\"><span style=\"font-weight: bold\"> Non-trainable params: </span><span style=\"color: #00af00; text-decoration-color: #00af00\">125,207</span> (489.09 KB)\n",
781
- "</pre>\n"
782
- ]
783
- },
784
- "metadata": {}
785
- }
786
- ]
787
- },
788
- {
789
- "cell_type": "code",
790
- "source": [
791
- "# Setup callbacks\n",
792
- "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n",
793
- "\n",
794
- "\n",
795
- "\n",
796
- "# Model checkpoint\n",
797
- "checkpoint_path = \"/content/drive/MyDrive/NTI-PRJCT/model_epoch_{epoch:02d}.keras\"\n",
798
- "checkpoint_callback = ModelCheckpoint(\n",
799
- " filepath=checkpoint_path,\n",
800
- " monitor='val_accuracy',\n",
801
- " save_best_only=False,\n",
802
- " save_weights_only=False,\n",
803
- " verbose=1\n",
804
- ")\n",
805
- "\n",
806
- "# Early stopping with proper patience\n",
807
- "early_stopping = EarlyStopping(\n",
808
- " monitor='val_loss',\n",
809
- " patience=8,\n",
810
- " restore_best_weights=True,\n",
811
- " verbose=1\n",
812
- ")"
813
- ],
814
- "metadata": {
815
- "id": "TUVta5W5IcTM"
816
- },
817
- "execution_count": null,
818
- "outputs": []
819
- },
820
- {
821
- "cell_type": "code",
822
- "source": [
823
- "lr_reducer = ReduceLROnPlateau(\n",
824
- " monitor='val_loss',\n",
825
- " factor=0.2,\n",
826
- " patience=5,\n",
827
- " min_lr=1e-7,\n",
828
- " verbose=1\n",
829
- ")\n",
830
- "\n",
831
- "# Training\n",
832
- "history = model.fit(\n",
833
- " train_ds,\n",
834
- " validation_data=val_ds,\n",
835
- " epochs=50,\n",
836
- " callbacks=[early_stopping, checkpoint_callback, lr_reducer],\n",
837
- " verbose=1\n",
838
- ")"
839
- ],
840
- "metadata": {
841
- "colab": {
842
- "base_uri": "https://localhost:8080/"
843
- },
844
- "id": "3jaEVVMHInmc",
845
- "outputId": "4420da4b-569b-43f3-fe36-734dab972380"
846
- },
847
- "execution_count": null,
848
- "outputs": [
849
- {
850
- "output_type": "stream",
851
- "name": "stdout",
852
- "text": [
853
- "Epoch 1/50\n",
854
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 857ms/step - accuracy: 0.7771 - loss: 0.4718\n",
855
- "Epoch 1: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_01.keras\n",
856
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m717s\u001b[0m 896ms/step - accuracy: 0.7772 - loss: 0.4718 - val_accuracy: 0.8034 - val_loss: 0.4270 - learning_rate: 0.0010\n",
857
- "Epoch 2/50\n",
858
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 857ms/step - accuracy: 0.8216 - loss: 0.3945\n",
859
- "Epoch 2: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_02.keras\n",
860
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m710s\u001b[0m 887ms/step - accuracy: 0.8216 - loss: 0.3945 - val_accuracy: 0.8350 - val_loss: 0.3646 - learning_rate: 0.0010\n",
861
- "Epoch 3/50\n",
862
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.8460 - loss: 0.3416\n",
863
- "Epoch 3: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_03.keras\n",
864
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m740s\u001b[0m 885ms/step - accuracy: 0.8460 - loss: 0.3416 - val_accuracy: 0.8519 - val_loss: 0.3786 - learning_rate: 0.0010\n",
865
- "Epoch 4/50\n",
866
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 857ms/step - accuracy: 0.8644 - loss: 0.3068\n",
867
- "Epoch 4: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_04.keras\n",
868
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m743s\u001b[0m 886ms/step - accuracy: 0.8644 - loss: 0.3068 - val_accuracy: 0.8444 - val_loss: 0.3616 - learning_rate: 0.0010\n",
869
- "Epoch 5/50\n",
870
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.8799 - loss: 0.2816\n",
871
- "Epoch 5: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_05.keras\n",
872
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m740s\u001b[0m 885ms/step - accuracy: 0.8799 - loss: 0.2816 - val_accuracy: 0.8388 - val_loss: 0.4151 - learning_rate: 0.0010\n",
873
- "Epoch 6/50\n",
874
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 855ms/step - accuracy: 0.8885 - loss: 0.2567\n",
875
- "Epoch 6: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_06.keras\n",
876
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m742s\u001b[0m 885ms/step - accuracy: 0.8885 - loss: 0.2567 - val_accuracy: 0.8512 - val_loss: 0.3529 - learning_rate: 0.0010\n",
877
- "Epoch 7/50\n",
878
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.8985 - loss: 0.2374\n",
879
- "Epoch 7: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_07.keras\n",
880
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m743s\u001b[0m 886ms/step - accuracy: 0.8985 - loss: 0.2374 - val_accuracy: 0.8609 - val_loss: 0.3472 - learning_rate: 0.0010\n",
881
- "Epoch 8/50\n",
882
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.9127 - loss: 0.2133\n",
883
- "Epoch 8: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_08.keras\n",
884
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m742s\u001b[0m 885ms/step - accuracy: 0.9127 - loss: 0.2133 - val_accuracy: 0.8087 - val_loss: 0.4966 - learning_rate: 0.0010\n",
885
- "Epoch 9/50\n",
886
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.9165 - loss: 0.2004\n",
887
- "Epoch 9: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_09.keras\n",
888
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m743s\u001b[0m 887ms/step - accuracy: 0.9165 - loss: 0.2004 - val_accuracy: 0.8619 - val_loss: 0.3643 - learning_rate: 0.0010\n",
889
- "Epoch 10/50\n",
890
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.9204 - loss: 0.1923\n",
891
- "Epoch 10: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_10.keras\n",
892
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m741s\u001b[0m 885ms/step - accuracy: 0.9204 - loss: 0.1923 - val_accuracy: 0.8597 - val_loss: 0.3573 - learning_rate: 0.0010\n",
893
- "Epoch 11/50\n",
894
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 855ms/step - accuracy: 0.9288 - loss: 0.1784\n",
895
- "Epoch 11: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_11.keras\n",
896
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m709s\u001b[0m 885ms/step - accuracy: 0.9288 - loss: 0.1784 - val_accuracy: 0.8697 - val_loss: 0.3494 - learning_rate: 0.0010\n",
897
- "Epoch 12/50\n",
898
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 854ms/step - accuracy: 0.9316 - loss: 0.1672\n",
899
- "Epoch 12: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_12.keras\n",
900
- "\n",
901
- "Epoch 12: ReduceLROnPlateau reducing learning rate to 0.00020000000949949026.\n",
902
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m706s\u001b[0m 881ms/step - accuracy: 0.9317 - loss: 0.1672 - val_accuracy: 0.8353 - val_loss: 0.4639 - learning_rate: 0.0010\n",
903
- "Epoch 13/50\n",
904
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 855ms/step - accuracy: 0.9546 - loss: 0.1189\n",
905
- "Epoch 13: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_13.keras\n",
906
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m709s\u001b[0m 885ms/step - accuracy: 0.9546 - loss: 0.1189 - val_accuracy: 0.9131 - val_loss: 0.2431 - learning_rate: 2.0000e-04\n",
907
- "Epoch 14/50\n",
908
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 855ms/step - accuracy: 0.9681 - loss: 0.0799\n",
909
- "Epoch 14: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_14.keras\n",
910
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m741s\u001b[0m 884ms/step - accuracy: 0.9681 - loss: 0.0799 - val_accuracy: 0.9187 - val_loss: 0.2386 - learning_rate: 2.0000e-04\n",
911
- "Epoch 15/50\n",
912
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.9742 - loss: 0.0689\n",
913
- "Epoch 15: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_15.keras\n",
914
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m708s\u001b[0m 885ms/step - accuracy: 0.9742 - loss: 0.0689 - val_accuracy: 0.9162 - val_loss: 0.2737 - learning_rate: 2.0000e-04\n",
915
- "Epoch 16/50\n",
916
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.9768 - loss: 0.0594\n",
917
- "Epoch 16: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_16.keras\n",
918
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m746s\u001b[0m 890ms/step - accuracy: 0.9768 - loss: 0.0594 - val_accuracy: 0.9106 - val_loss: 0.2691 - learning_rate: 2.0000e-04\n",
919
- "Epoch 17/50\n",
920
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━��━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 857ms/step - accuracy: 0.9784 - loss: 0.0552\n",
921
- "Epoch 17: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_17.keras\n",
922
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m744s\u001b[0m 893ms/step - accuracy: 0.9784 - loss: 0.0552 - val_accuracy: 0.9147 - val_loss: 0.2741 - learning_rate: 2.0000e-04\n",
923
- "Epoch 18/50\n",
924
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 858ms/step - accuracy: 0.9805 - loss: 0.0505\n",
925
- "Epoch 18: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_18.keras\n",
926
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m716s\u001b[0m 895ms/step - accuracy: 0.9805 - loss: 0.0505 - val_accuracy: 0.9184 - val_loss: 0.2681 - learning_rate: 2.0000e-04\n",
927
- "Epoch 19/50\n",
928
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 861ms/step - accuracy: 0.9843 - loss: 0.0412\n",
929
- "Epoch 19: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_19.keras\n",
930
- "\n",
931
- "Epoch 19: ReduceLROnPlateau reducing learning rate to 4.0000001899898055e-05.\n",
932
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m739s\u001b[0m 891ms/step - accuracy: 0.9843 - loss: 0.0412 - val_accuracy: 0.9159 - val_loss: 0.2972 - learning_rate: 2.0000e-04\n",
933
- "Epoch 20/50\n",
934
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 856ms/step - accuracy: 0.9862 - loss: 0.0367\n",
935
- "Epoch 20: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_20.keras\n",
936
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m711s\u001b[0m 888ms/step - accuracy: 0.9862 - loss: 0.0367 - val_accuracy: 0.9156 - val_loss: 0.3108 - learning_rate: 4.0000e-05\n",
937
- "Epoch 21/50\n",
938
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m0s\u001b[0m 860ms/step - accuracy: 0.9897 - loss: 0.0274\n",
939
- "Epoch 21: saving model to /content/drive/MyDrive/NTI-PRJCT/model_epoch_21.keras\n",
940
- "\u001b[1m800/800\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m716s\u001b[0m 894ms/step - accuracy: 0.9897 - loss: 0.0274 - val_accuracy: 0.9184 - val_loss: 0.3099 - learning_rate: 4.0000e-05\n",
941
- "Epoch 22/50\n",
942
- "\u001b[1m150/800\u001b[0m \u001b[32m━━━\u001b[0m\u001b[37m━━━━━━━━━━━━━━━━━\u001b[0m \u001b[1m9:21\u001b[0m 864ms/step - accuracy: 0.9879 - loss: 0.0310"
943
- ]
944
- }
945
- ]
946
- },
947
- {
948
- "cell_type": "code",
949
- "source": [],
950
- "metadata": {
951
- "id": "KFBcgWXf3kLt"
952
- },
953
- "execution_count": null,
954
- "outputs": []
955
- }
956
- ]
957
- }