diff --git "a/src/TrashSort_Training_Notebook.ipynb" "b/src/TrashSort_Training_Notebook.ipynb" new file mode 100644--- /dev/null +++ "b/src/TrashSort_Training_Notebook.ipynb" @@ -0,0 +1,336 @@ +{ + "cells": [ + { + "cell_type": "code", + "execution_count": 8, + "id": "ac80a687", + "metadata": {}, + "outputs": [], + "source": [ + "import tensorflow as tf\n", + "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n", + "from tensorflow.keras.applications import MobileNetV2\n", + "from tensorflow.keras.layers import Dense, GlobalAveragePooling2D, Dropout\n", + "from tensorflow.keras.models import Model\n", + "from tensorflow.keras.optimizers import Adam\n", + "from sklearn.metrics import confusion_matrix, classification_report\n", + "import matplotlib.pyplot as plt\n", + "import seaborn as sns\n", + "import numpy as np\n", + "import os" + ] + }, + { + "cell_type": "code", + "execution_count": 9, + "id": "e45a6baa", + "metadata": {}, + "outputs": [], + "source": [ + "# Parameters\n", + "IMG_SIZE = 128\n", + "BATCH_SIZE = 32\n", + "EPOCHS = 15\n", + "DATA_DIR = r\"C:\\Users\\Cherilyn\\Desktop\\trash_sort\\src\\data\\trashnet\"\n", + "MODEL_PATH = 'models/trashsort_cnn.h5'" + ] + }, + { + "cell_type": "code", + "execution_count": 10, + "id": "957a1e3e", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Current working directory: c:\\Users\\Cherilyn\\Desktop\\trash_sort\\src\n" + ] + } + ], + "source": [ + "# Verify current working directory (optional)\n", + "print(\"Current working directory:\", os.getcwd())" + ] + }, + { + "cell_type": "code", + "execution_count": 11, + "id": "2213192b", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Found 1914 images belonging to 5 classes.\n", + "Found 476 images belonging to 5 classes.\n", + "Detected classes: ['cardboard', 'glass', 'metal', 'paper', 'plastic']\n" + ] + } + ], + "source": [ + "# Data generators\n", + "train_datagen = ImageDataGenerator(\n", + " rescale=1./255,\n", + " validation_split=0.2,\n", + " horizontal_flip=True,\n", + " rotation_range=15\n", + ")\n", + "\n", + "train_gen = train_datagen.flow_from_directory(\n", + " DATA_DIR,\n", + " target_size=(IMG_SIZE, IMG_SIZE),\n", + " batch_size=BATCH_SIZE,\n", + " class_mode='categorical',\n", + " subset='training',\n", + " shuffle=True\n", + ")\n", + "\n", + "val_gen = train_datagen.flow_from_directory(\n", + " DATA_DIR,\n", + " target_size=(IMG_SIZE, IMG_SIZE),\n", + " batch_size=BATCH_SIZE,\n", + " class_mode='categorical',\n", + " subset='validation',\n", + " shuffle=False\n", + ")\n", + "\n", + "# Get class names from generator\n", + "CLASS_NAMES = list(train_gen.class_indices.keys())\n", + "print(\"Detected classes:\", CLASS_NAMES)" + ] + }, + { + "cell_type": "code", + "execution_count": 12, + "id": "7a66c30c", + "metadata": {}, + "outputs": [], + "source": [ + "# Build model using MobileNetV2\n", + "base_model = MobileNetV2(weights='imagenet', include_top=False, input_shape=(IMG_SIZE, IMG_SIZE, 3))\n", + "base_model.trainable = False # Freeze base model\n", + "\n", + "x = base_model.output\n", + "x = GlobalAveragePooling2D()(x)\n", + "x = Dropout(0.3)(x)\n", + "predictions = Dense(len(CLASS_NAMES), activation='softmax')(x)\n", + "\n", + "model = Model(inputs=base_model.input, outputs=predictions)\n", + "model.compile(optimizer=Adam(learning_rate=1e-3), loss='categorical_crossentropy', metrics=['accuracy'])" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "id": "8c4c5d15", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "c:\\Users\\Cherilyn\\AppData\\Local\\Programs\\Python\\Python312\\Lib\\site-packages\\keras\\src\\trainers\\data_adapters\\py_dataset_adapter.py:121: UserWarning: Your `PyDataset` class should call `super().__init__(**kwargs)` in its constructor. `**kwargs` can include `workers`, `use_multiprocessing`, `max_queue_size`. Do not pass these arguments to `fit()`, as they will be ignored.\n", + " self._warn_if_super_not_called()\n" + ] + }, + { + "name": "stdout", + "output_type": "stream", + "text": [ + "Epoch 1/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m37s\u001b[0m 562ms/step - accuracy: 0.3658 - loss: 1.6455 - val_accuracy: 0.6303 - val_loss: 0.9275\n", + "Epoch 2/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 272ms/step - accuracy: 0.6936 - loss: 0.7798 - val_accuracy: 0.6765 - val_loss: 0.8273\n", + "Epoch 3/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 273ms/step - accuracy: 0.7771 - loss: 0.6054 - val_accuracy: 0.7290 - val_loss: 0.7569\n", + "Epoch 4/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 269ms/step - accuracy: 0.8023 - loss: 0.5575 - val_accuracy: 0.7059 - val_loss: 0.7802\n", + "Epoch 5/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 272ms/step - accuracy: 0.8167 - loss: 0.4857 - val_accuracy: 0.7290 - val_loss: 0.7199\n", + "Epoch 6/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 270ms/step - accuracy: 0.8273 - loss: 0.4506 - val_accuracy: 0.7626 - val_loss: 0.6881\n", + "Epoch 7/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 269ms/step - accuracy: 0.8527 - loss: 0.3887 - val_accuracy: 0.7332 - val_loss: 0.7583\n", + "Epoch 8/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 271ms/step - accuracy: 0.8408 - loss: 0.4061 - val_accuracy: 0.7437 - val_loss: 0.7043\n", + "Epoch 9/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 272ms/step - accuracy: 0.8605 - loss: 0.3765 - val_accuracy: 0.7479 - val_loss: 0.7031\n", + "Epoch 10/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 270ms/step - accuracy: 0.8761 - loss: 0.3413 - val_accuracy: 0.7605 - val_loss: 0.7077\n", + "Epoch 11/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 275ms/step - accuracy: 0.8923 - loss: 0.3267 - val_accuracy: 0.7563 - val_loss: 0.6693\n", + "Epoch 12/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 269ms/step - accuracy: 0.8753 - loss: 0.3337 - val_accuracy: 0.7752 - val_loss: 0.6822\n", + "Epoch 13/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 271ms/step - accuracy: 0.8866 - loss: 0.3294 - val_accuracy: 0.7479 - val_loss: 0.6629\n", + "Epoch 14/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m17s\u001b[0m 277ms/step - accuracy: 0.8790 - loss: 0.3133 - val_accuracy: 0.7647 - val_loss: 0.6745\n", + "Epoch 15/15\n", + "\u001b[1m60/60\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m16s\u001b[0m 272ms/step - accuracy: 0.9020 - loss: 0.2824 - val_accuracy: 0.7626 - val_loss: 0.7227\n" + ] + } + ], + "source": [ + "# Train model\n", + "history = model.fit(\n", + " train_gen,\n", + " validation_data=val_gen,\n", + " epochs=EPOCHS\n", + ")" + ] + }, + { + "cell_type": "code", + "execution_count": 14, + "id": "c16f29c9", + "metadata": {}, + "outputs": [ + { + "name": "stderr", + "output_type": "stream", + "text": [ + "WARNING:absl:You are saving your model as an HDF5 file via `model.save()` or `keras.saving.save_model(model)`. This file format is considered legacy. We recommend using instead the native Keras format, e.g. `model.save('my_model.keras')` or `keras.saving.save_model(model, 'my_model.keras')`. \n" + ] + } + ], + "source": [ + "# Save model\n", + "os.makedirs(os.path.dirname(MODEL_PATH), exist_ok=True)\n", + "model.save(MODEL_PATH)" + ] + }, + { + "cell_type": "code", + "execution_count": 15, + "id": "241fc8ea", + "metadata": {}, + "outputs": [ + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Plot accuracy and loss\n", + "plt.figure(figsize=(12, 5))\n", + "\n", + "plt.subplot(1, 2, 1)\n", + "plt.plot(history.history['accuracy'], label='Train Acc')\n", + "plt.plot(history.history['val_accuracy'], label='Val Acc')\n", + "plt.legend()\n", + "plt.title('Accuracy')\n", + "\n", + "plt.subplot(1, 2, 2)\n", + "plt.plot(history.history['loss'], label='Train Loss')\n", + "plt.plot(history.history['val_loss'], label='Val Loss')\n", + "plt.legend()\n", + "plt.title('Loss')\n", + "\n", + "plt.savefig('training_history.png')\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 16, + "id": "c021009a", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "\u001b[1m15/15\u001b[0m \u001b[32m━━━━━━━━━━━━━━━━━━━━\u001b[0m\u001b[37m\u001b[0m \u001b[1m5s\u001b[0m 284ms/step\n" + ] + }, + { + "data": { + "image/png": "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", + "text/plain": [ + "
" + ] + }, + "metadata": {}, + "output_type": "display_data" + } + ], + "source": [ + "# Confusion Matrix\n", + "val_gen.reset()\n", + "Y_pred = model.predict(val_gen, verbose=1)\n", + "y_pred = np.argmax(Y_pred, axis=1)\n", + "y_true = val_gen.classes\n", + "\n", + "cm = confusion_matrix(y_true, y_pred)\n", + "plt.figure(figsize=(8, 6))\n", + "sns.heatmap(cm, annot=True, fmt='d', cmap='Blues', xticklabels=CLASS_NAMES, yticklabels=CLASS_NAMES)\n", + "plt.title('Confusion Matrix')\n", + "plt.xlabel('Predicted')\n", + "plt.ylabel('Actual')\n", + "plt.tight_layout()\n", + "plt.savefig('confusion_matrix.png')\n", + "plt.show()" + ] + }, + { + "cell_type": "code", + "execution_count": 17, + "id": "4a2c4514", + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + " precision recall f1-score support\n", + "\n", + " cardboard 0.95 0.66 0.78 80\n", + " glass 0.69 0.68 0.68 100\n", + " metal 0.73 0.74 0.74 82\n", + " paper 0.72 0.95 0.82 118\n", + " plastic 0.77 0.67 0.72 96\n", + "\n", + " accuracy 0.75 476\n", + " macro avg 0.77 0.74 0.75 476\n", + "weighted avg 0.76 0.75 0.75 476\n", + "\n" + ] + } + ], + "source": [ + "# Print classification report\n", + "print(classification_report(y_true, y_pred, target_names=CLASS_NAMES))" + ] + } + ], + "metadata": { + "kernelspec": { + "display_name": "Python 3", + "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": "3.12.2" + } + }, + "nbformat": 4, + "nbformat_minor": 5 +}