{
"cells": [
{
"cell_type": "code",
"execution_count": 51,
"id": "62c6c5c3-d2ba-4e79-b533-828c3083e8ea",
"metadata": {
"is_executing": true
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Num GPUs Available: 1\n",
"Physical devices cannot be modified after being initialized\n",
"Found 1529 images belonging to 6 classes.\n",
"Found 380 images belonging to 6 classes.\n",
"Epoch 1/10\n",
"24/24 [==============================] - 21s 851ms/step - loss: 1.1108 - accuracy: 0.5134 - val_loss: 0.8370 - val_accuracy: 0.6842\n",
"Epoch 2/10\n",
"24/24 [==============================] - 21s 862ms/step - loss: 0.6156 - accuracy: 0.7613 - val_loss: 0.4063 - val_accuracy: 0.8447\n",
"Epoch 3/10\n",
"24/24 [==============================] - 22s 915ms/step - loss: 0.2704 - accuracy: 0.9006 - val_loss: 0.2283 - val_accuracy: 0.9316\n",
"Epoch 4/10\n",
"24/24 [==============================] - 22s 909ms/step - loss: 0.1532 - accuracy: 0.9477 - val_loss: 0.1145 - val_accuracy: 0.9579\n",
"Epoch 5/10\n",
"24/24 [==============================] - 22s 905ms/step - loss: 0.1324 - accuracy: 0.9516 - val_loss: 0.2525 - val_accuracy: 0.9026\n",
"Epoch 6/10\n",
"24/24 [==============================] - 22s 912ms/step - loss: 0.0737 - accuracy: 0.9719 - val_loss: 0.0519 - val_accuracy: 0.9816\n",
"Epoch 7/10\n",
"24/24 [==============================] - 23s 952ms/step - loss: 0.0663 - accuracy: 0.9797 - val_loss: 0.0544 - val_accuracy: 0.9842\n",
"Epoch 8/10\n",
"24/24 [==============================] - 22s 918ms/step - loss: 0.0482 - accuracy: 0.9817 - val_loss: 0.0489 - val_accuracy: 0.9711\n",
"Epoch 9/10\n",
"24/24 [==============================] - 22s 928ms/step - loss: 0.0240 - accuracy: 0.9908 - val_loss: 0.0337 - val_accuracy: 0.9868\n",
"Epoch 10/10\n",
"24/24 [==============================] - 22s 911ms/step - loss: 0.0224 - accuracy: 0.9922 - val_loss: 0.0411 - val_accuracy: 0.9895\n",
"6/6 [==============================] - 4s 717ms/step - loss: 0.0078 - accuracy: 1.0000\n",
"Test loss: 0.007844111882150173, Test accuracy: 1.0\n",
"Model: \"sequential_32\"\n",
"_________________________________________________________________\n",
" Layer (type) Output Shape Param # \n",
"=================================================================\n",
" conv2d_128 (Conv2D) (None, 222, 222, 32) 896 \n",
" \n",
" max_pooling2d_128 (MaxPooli (None, 111, 111, 32) 0 \n",
" ng2D) \n",
" \n",
" conv2d_129 (Conv2D) (None, 109, 109, 64) 18496 \n",
" \n",
" max_pooling2d_129 (MaxPooli (None, 54, 54, 64) 0 \n",
" ng2D) \n",
" \n",
" conv2d_130 (Conv2D) (None, 52, 52, 128) 73856 \n",
" \n",
" max_pooling2d_130 (MaxPooli (None, 26, 26, 128) 0 \n",
" ng2D) \n",
" \n",
" conv2d_131 (Conv2D) (None, 24, 24, 256) 295168 \n",
" \n",
" max_pooling2d_131 (MaxPooli (None, 12, 12, 256) 0 \n",
" ng2D) \n",
" \n",
" flatten_32 (Flatten) (None, 36864) 0 \n",
" \n",
" dense_96 (Dense) (None, 256) 9437440 \n",
" \n",
" dense_97 (Dense) (None, 128) 32896 \n",
" \n",
" dense_98 (Dense) (None, 6) 774 \n",
" \n",
"=================================================================\n",
"Total params: 9,859,526\n",
"Trainable params: 9,859,526\n",
"Non-trainable params: 0\n",
"_________________________________________________________________\n",
"It took: 3.74 minutes\n"
]
}
],
"source": [
"import time\n",
"from keras.models import Sequential\n",
"from keras.layers import Conv2D, MaxPooling2D, Flatten, Dense\n",
"from keras.preprocessing.image import ImageDataGenerator\n",
"from keras.losses import CategoricalCrossentropy\n",
"from tensorflow.keras.callbacks import EarlyStopping\n",
"from tensorflow.keras.optimizers import Adam\n",
"import tensorflow as tf\n",
"\n",
"from Func.getSubFolders import count_sub_folders\n",
"\n",
"\n",
"path = 'Data'\n",
"output = 'Model/keras_model.h5'\n",
"start_time = time.time()\n",
"\n",
"physical_devices = tf.config.list_physical_devices('GPU')\n",
"print(\"Num GPUs Available: \", len(physical_devices))\n",
"if len(physical_devices) > 0:\n",
" try:\n",
" # Set memory growth to true\n",
" for device in physical_devices:\n",
" tf.config.experimental.set_memory_growth(device, True)\n",
" \n",
" # Optionally, you can set a memory limit if necessary\n",
" # tf.config.set_logical_device_configuration(\n",
" # physical_devices[0],\n",
" # [tf.config.LogicalDeviceConfiguration(memory_limit=4096)]) # Set to 4GB\n",
" except RuntimeError as e:\n",
" print(e)\n",
"else:\n",
" print(\"No GPU available. Using CPU.\")\n",
"\n",
"\n",
"# Step 1: Load and Preprocess Images\n",
"datagen = ImageDataGenerator(\n",
" rescale=1. / 255,\n",
" validation_split=0.2,\n",
" width_shift_range=0.2,\n",
" height_shift_range=0.2,\n",
" shear_range=0.2,\n",
" zoom_range=0.2,\n",
")\n",
"\n",
"test_datagen = ImageDataGenerator(rescale=1. / 255)\n",
"\n",
"\n",
"# Step 2: Label the Data\n",
"train_set = datagen.flow_from_directory(\n",
" path,\n",
" target_size=(300, 300),\n",
" batch_size=64,\n",
" class_mode='categorical',\n",
" subset='training'\n",
")\n",
"\n",
"test_set = datagen.flow_from_directory(\n",
" path,\n",
" target_size=(300, 300),\n",
" batch_size=64,\n",
" class_mode='categorical',\n",
" subset='validation'\n",
")\n",
"\n",
"# draw Images\n",
"# images, labels = next(test_set)\n",
"\n",
"# class_indices = test_set.class_indices\n",
"# plot_images(images, labels, class_indices)\n",
"\n",
"\n",
"\n",
"# Step 4: Build the Model\n",
"model = Sequential()\n",
"model.add(Conv2D(32, (3, 3), input_shape=(300, 300, 3), activation='relu'))\n",
"model.add(MaxPooling2D(pool_size=(2, 2)))\n",
"\n",
"model.add(Conv2D(64, (3, 3), activation='relu'))\n",
"model.add(MaxPooling2D((2, 2)))\n",
"\n",
"model.add(Conv2D(128, (3, 3), activation='relu'))\n",
"model.add(MaxPooling2D((2, 2)))\n",
"\n",
"model.add(Conv2D(256, (3, 3), activation='relu'))\n",
"model.add(MaxPooling2D((2, 2)))\n",
"\n",
"model.add(Flatten())\n",
"model.add(Dense(units=256, activation='relu'))\n",
"model.add(Dense(units=128, activation='relu'))\n",
"model.add(Dense(units=count_sub_folders(path), activation='softmax'))\n",
"# def createLayers(input_shape=(224, 224, 3)):\n",
"# inputs = tf.keras.Input(shape=input_shape)\n",
"#\n",
"# x = layers.Conv2D(48, (1, 1), padding='same', use_bias=False, name='block_1_expand')(inputs)\n",
"# x = layers.BatchNormalization(name='block_1_expand_BN')(x)\n",
"# x = layers.ReLU(6., name='block_1_expand_relu')(x)\n",
"#\n",
"# x = layers.DepthwiseConv2D((3, 3), padding='same', use_bias=False, name='block_1_depthwise')(x)\n",
"# x = layers.BatchNormalization(name='block_1_depthwise_BN')(x)\n",
"# x = layers.ReLU(6., name='block_1_depthwise_relu')(x)\n",
"#\n",
"# x = layers.Conv2D(8, (1, 1), padding='same', use_bias=False, name='block_1_project')(x)\n",
"# x = layers.BatchNormalization(name='block_1_project_BN')(x)\n",
"#\n",
"# for i in range(2,5):\n",
"# x1 = layers.Conv2D(48, (1, 1), padding='same', use_bias=False, name=f'block_{i}_expand')(x)\n",
"# x1 = layers.BatchNormalization(name=f'block_{i}_expand_BN')(x1)\n",
"# x1 = layers.ReLU(6., name=f'block_{i}_expand_relu')(x1)\n",
"#\n",
"# x1 = layers.DepthwiseConv2D((3, 3), padding='same', use_bias=False, name=f'block_{i}_depthwise')(x1)\n",
"# x1 = layers.BatchNormalization(name=f'block_{i}_depthwise_BN')(x1)\n",
"# x1 = layers.ReLU(6., name=f'block_{i}_depthwise_relu')(x1)\n",
"#\n",
"# x1 = layers.Conv2D(8, (1, 1), padding='same', use_bias=False, name=f'block_{i}_project')(x1)\n",
"# x1 = layers.BatchNormalization(name=f'block_{i}_project_BN')(x1)\n",
"#\n",
"# x = layers.Add(name=f'block_{i}_add')([x, x1])\n",
"#\n",
"# x = tf.keras.layers.GlobalAveragePooling2D()(x)\n",
"# outputs = tf.keras.layers.Dense(count_sub_folders(path), activation='softmax')(x)\n",
"# model = models.Model(inputs, outputs, name='testModel')\n",
"#\n",
"# return model\n",
"#\n",
"#\n",
"# model = createLayers()\n",
"\n",
"# Compile the Model after pruning\n",
"model.compile(optimizer=Adam(learning_rate=0.001),\n",
" loss=CategoricalCrossentropy(from_logits=False),\n",
" metrics=['accuracy'])\n",
"\n",
"# Step 6: Train the Model\n",
"early_stopping = EarlyStopping(monitor='val_loss', patience=5, restore_best_weights=True)\n",
"\n",
"model.fit(train_set, validation_data=test_set, epochs=10, callbacks=early_stopping)\n",
"\n",
"# Step 7: Evaluate the Model\n",
"loss, accuracy = model.evaluate(test_set)\n",
"print(f'Test loss: {loss}, Test accuracy: {accuracy}')\n",
"\n",
"# Save the trained model\n",
"model.save(output)\n",
"end_time = time.time()\n",
"\n",
"execute_time = (end_time - start_time) / 60\n",
"\n",
"model.summary()\n",
"\n",
"# Print the result\n",
"print(f\"It took: {execute_time:0.2f} minutes\")\n"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "ec102c63-b7d9-4f22-9f3e-a6bd12995a4f",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.9.19"
}
},
"nbformat": 4,
"nbformat_minor": 5
}