{ "cells": [ { "cell_type": "markdown", "id": "983a6893-ecf1-494a-bcf0-3da2ddf4f550", "metadata": {}, "source": [ "### CNN DateFruit Classification" ] }, { "cell_type": "markdown", "id": "876c776a-bb81-4141-b1f7-838fe0374ebb", "metadata": {}, "source": [ "" ] }, { "cell_type": "markdown", "id": "8b57ea2a-cfcb-4338-b643-cfe9eaf728c3", "metadata": {}, "source": [ "## Introduction\n", "The goal of the model was to recognize different types of dates from images. I built my own CNN model without using pre-trained models (no transfer learning). I wanted to show that the model can work well even when fully created from scratch." ] }, { "cell_type": "code", "execution_count": 1, "id": "17dd65b6-0fd4-49c8-b508-1169a08d0f56", "metadata": {}, "outputs": [], "source": [ "#pip install kagglehub" ] }, { "cell_type": "code", "execution_count": 1, "id": "598deb58-58aa-4563-8139-f4ee04b2fb0b", "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ "/usr/local/lib/python3.12/dist-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n", " from .autonotebook import tqdm as notebook_tqdm\n" ] } ], "source": [ "import kagglehub\n", "\n", "# Download latest version\n", "path = kagglehub.dataset_download(\"wadhasnalhamdan/date-fruit-image-dataset-in-controlled-environment\")" ] }, { "cell_type": "code", "execution_count": 2, "id": "66921412-5efa-4723-a46c-8f44172f1b99", "metadata": {}, "outputs": [], "source": [ "import os\n", "\n", "os.environ['CUDA_DEVICE_ORDER'] = 'PCI_BUS_ID'\n", "os.environ['CUDA_LAUNCH_BLOCKING'] = '0' # Delay-Timing\n", "os.environ['TF_CPP_MIN_LOG_LEVEL'] = '3' # TensorFlow\n", "os.environ['NVIDIA_TF32_OVERRIDE'] = '0' # TF32 Profiling" ] }, { "cell_type": "code", "execution_count": 3, "id": "12245595-aaa6-480a-8356-c44c3c20399a", "metadata": {}, "outputs": [], "source": [ "import tensorflow as tf\n", "tf.debugging.disable_traceback_filtering()" ] }, { "cell_type": "code", "execution_count": 4, "id": "6dec7e2d-4fac-480d-9d4c-00231e64158b", "metadata": {}, "outputs": [], "source": [ "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n", "from tensorflow.keras.models import Sequential\n", "from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D, Flatten, Dense, Dropout, BatchNormalization\n", "from tensorflow.keras.optimizers import Adam\n", "from tensorflow.keras.callbacks import EarlyStopping, ModelCheckpoint, ReduceLROnPlateau\n", "from tensorflow.keras.layers import GlobalAveragePooling2D\n", "import matplotlib.pyplot as plt\n", "from sklearn.metrics import confusion_matrix, classification_report\n", "import seaborn as sns\n", "import numpy as np\n", "import tensorflow as tf\n", "import pickle\n", "import warnings\n", "warnings.filterwarnings(\"ignore\") " ] }, { "cell_type": "code", "execution_count": 5, "id": "a9edd8a3-8ac5-46dd-bcb0-6eb71e935121", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Found 1329 images belonging to 9 classes.\n", "Found 329 images belonging to 9 classes.\n" ] } ], "source": [ "data_dir = path\n", "\n", "from tensorflow.keras.preprocessing.image import ImageDataGenerator\n", "\n", "img_size = (224, 224)\n", "batch_size = 16 # oder 32, aber 16 ist oft etwas besser für val_loss\n", "\n", "train_datagen = ImageDataGenerator(\n", " rescale=1./255,\n", " validation_split=0.2,\n", " rotation_range=25,\n", " width_shift_range=0.15,\n", " height_shift_range=0.15,\n", " zoom_range=0.15,\n", " shear_range=0.15,\n", " horizontal_flip=True,\n", " fill_mode='nearest'\n", ")\n", "\n", "val_datagen = ImageDataGenerator(\n", " rescale=1./255,\n", " validation_split=0.2\n", ")\n", "\n", "train_data = train_datagen.flow_from_directory(\n", " data_dir,\n", " target_size=img_size,\n", " batch_size=batch_size,\n", " subset='training',\n", " class_mode='sparse',\n", " shuffle=True\n", ")\n", "\n", "val_data = val_datagen.flow_from_directory(\n", " data_dir,\n", " target_size=img_size,\n", " batch_size=batch_size,\n", " subset='validation',\n", " class_mode='sparse',\n", " shuffle=False\n", ")" ] }, { "cell_type": "code", "execution_count": 42, "id": "d5cf3614-5e9d-4122-be81-6bd1b2436f83", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Model: \"sequential_6\"\n", "_________________________________________________________________\n", " Layer (type) Output Shape Param # \n", "=================================================================\n", " conv2d_23 (Conv2D) (None, 224, 224, 32) 896 \n", " \n", " conv2d_24 (Conv2D) (None, 224, 224, 32) 9248 \n", " \n", " max_pooling2d_16 (MaxPooli (None, 112, 112, 32) 0 \n", " ng2D) \n", " \n", " conv2d_25 (Conv2D) (None, 112, 112, 64) 18496 \n", " \n", " conv2d_26 (Conv2D) (None, 112, 112, 64) 36928 \n", " \n", " max_pooling2d_17 (MaxPooli (None, 56, 56, 64) 0 \n", " ng2D) \n", " \n", " conv2d_27 (Conv2D) (None, 56, 56, 128) 73856 \n", " \n", " max_pooling2d_18 (MaxPooli (None, 28, 28, 128) 0 \n", " ng2D) \n", " \n", " flatten_5 (Flatten) (None, 100352) 0 \n", " \n", " dense_24 (Dense) (None, 256) 25690368 \n", " \n", " dropout_17 (Dropout) (None, 256) 0 \n", " \n", " dense_25 (Dense) (None, 9) 2313 \n", " \n", "=================================================================\n", "Total params: 25832105 (98.54 MB)\n", "Trainable params: 25832105 (98.54 MB)\n", "Non-trainable params: 0 (0.00 Byte)\n", "_________________________________________________________________\n" ] } ], "source": [ "model = Sequential()\n", "model.add(Input(shape=(224,224,3)))\n", "\n", "model.add(Conv2D(32, (3,3), activation='relu', padding='same'))\n", "model.add(Conv2D(32, (3,3), activation='relu', padding='same'))\n", "model.add(MaxPooling2D((2,2)))\n", "model.add(Conv2D(64, (3,3), activation='relu', padding='same'))\n", "model.add(Conv2D(64, (3,3), activation='relu', padding='same'))\n", "model.add(MaxPooling2D((2,2))) \n", "model.add(Conv2D(128, (3,3), activation='relu', padding='same'))\n", "model.add(MaxPooling2D((2,2)))\n", "model.add(Flatten())\n", "model.add(Dense(256, activation='relu'))\n", "model.add(Dropout(0.5))\n", "model.add(Dense(train_data.num_classes, activation='softmax'))\n", "\n", "model.compile(\n", " optimizer=Adam(learning_rate=5e-4),\n", " loss='sparse_categorical_crossentropy',\n", " metrics=['accuracy']\n", ")\n", "\n", "model.summary()" ] }, { "cell_type": "code", "execution_count": 43, "id": "9125069a-1246-4712-bd51-1aaa10ab1cb0", "metadata": {}, "outputs": [], "source": [ "early_stop = EarlyStopping(\n", " monitor='val_loss',\n", " patience=7,\n", " restore_best_weights=True\n", ")\n", "\n", "checkpoint = ModelCheckpoint(\n", " filepath='best_date_cnn.keras',\n", " monitor='val_loss',\n", " save_best_only=True\n", ")\n", "\n", "lr_schedule = ReduceLROnPlateau(\n", " monitor='val_loss',\n", " factor=0.5, \n", " patience=2, \n", " min_lr=1e-6,\n", " verbose=1\n", ")" ] }, { "cell_type": "code", "execution_count": 44, "id": "68bd83a0-3149-4a3c-8adb-953a3c5d8dad", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/40\n" ] }, { "name": "stderr", "output_type": "stream", "text": [ "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ " 1/84 [..............................] - ETA: 2:21 - loss: 2.1995 - accuracy: 0.0000e+00" ] }, { "name": "stderr", "output_type": "stream", "text": [ "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "84/84 [==============================] - 52s 605ms/step - loss: 2.2133 - accuracy: 0.1625 - val_loss: 2.0029 - val_accuracy: 0.2462 - lr: 5.0000e-04\n", "Epoch 2/40\n", "84/84 [==============================] - 50s 594ms/step - loss: 1.8961 - accuracy: 0.2769 - val_loss: 1.6982 - val_accuracy: 0.3222 - lr: 5.0000e-04\n", "Epoch 3/40\n", "84/84 [==============================] - 51s 609ms/step - loss: 1.6872 - accuracy: 0.3476 - val_loss: 1.5724 - val_accuracy: 0.4316 - lr: 5.0000e-04\n", "Epoch 4/40\n", "84/84 [==============================] - 50s 588ms/step - loss: 1.5899 - accuracy: 0.3958 - val_loss: 1.5368 - val_accuracy: 0.4529 - lr: 5.0000e-04\n", "Epoch 5/40\n", "84/84 [==============================] - 49s 587ms/step - loss: 1.5390 - accuracy: 0.4011 - val_loss: 1.3234 - val_accuracy: 0.5502 - lr: 5.0000e-04\n", "Epoch 6/40\n", "84/84 [==============================] - 49s 584ms/step - loss: 1.4844 - accuracy: 0.4417 - val_loss: 1.5232 - val_accuracy: 0.4438 - lr: 5.0000e-04\n", "Epoch 7/40\n", "84/84 [==============================] - 50s 596ms/step - loss: 1.4848 - accuracy: 0.4319 - val_loss: 1.1718 - val_accuracy: 0.5745 - lr: 5.0000e-04\n", "Epoch 8/40\n", "84/84 [==============================] - 50s 592ms/step - loss: 1.3790 - accuracy: 0.4620 - val_loss: 1.2224 - val_accuracy: 0.5684 - lr: 5.0000e-04\n", "Epoch 9/40\n", "84/84 [==============================] - 49s 580ms/step - loss: 1.3053 - accuracy: 0.5049 - val_loss: 1.1292 - val_accuracy: 0.5957 - lr: 5.0000e-04\n", "Epoch 10/40\n", "84/84 [==============================] - 49s 579ms/step - loss: 1.3088 - accuracy: 0.5019 - val_loss: 1.1574 - val_accuracy: 0.5805 - lr: 5.0000e-04\n", "Epoch 11/40\n", "84/84 [==============================] - ETA: 0s - loss: 1.2579 - accuracy: 0.5192 \n", "Epoch 11: ReduceLROnPlateau reducing learning rate to 0.0002500000118743628.\n", "84/84 [==============================] - 48s 570ms/step - loss: 1.2579 - accuracy: 0.5192 - val_loss: 1.1542 - val_accuracy: 0.6353 - lr: 5.0000e-04\n", "Epoch 12/40\n", "84/84 [==============================] - 50s 596ms/step - loss: 1.1559 - accuracy: 0.5628 - val_loss: 0.9113 - val_accuracy: 0.6900 - lr: 2.5000e-04\n", "Epoch 13/40\n", "84/84 [==============================] - 49s 581ms/step - loss: 1.0848 - accuracy: 0.6035 - val_loss: 1.0976 - val_accuracy: 0.6140 - lr: 2.5000e-04\n", "Epoch 14/40\n", "84/84 [==============================] - 49s 588ms/step - loss: 1.0622 - accuracy: 0.5989 - val_loss: 0.8911 - val_accuracy: 0.7082 - lr: 2.5000e-04\n", "Epoch 15/40\n", "84/84 [==============================] - 49s 583ms/step - loss: 1.0211 - accuracy: 0.6065 - val_loss: 1.0181 - val_accuracy: 0.6322 - lr: 2.5000e-04\n", "Epoch 16/40\n", "84/84 [==============================] - 50s 591ms/step - loss: 0.9352 - accuracy: 0.6494 - val_loss: 0.8806 - val_accuracy: 0.7082 - lr: 2.5000e-04\n", "Epoch 17/40\n", "84/84 [==============================] - 49s 579ms/step - loss: 0.9133 - accuracy: 0.6614 - val_loss: 0.8156 - val_accuracy: 0.7112 - lr: 2.5000e-04\n", "Epoch 18/40\n", "84/84 [==============================] - 50s 589ms/step - loss: 0.8630 - accuracy: 0.6840 - val_loss: 0.8845 - val_accuracy: 0.6930 - lr: 2.5000e-04\n", "Epoch 19/40\n", "84/84 [==============================] - 50s 598ms/step - loss: 0.8662 - accuracy: 0.7065 - val_loss: 0.6943 - val_accuracy: 0.7751 - lr: 2.5000e-04\n", "Epoch 20/40\n", "84/84 [==============================] - 50s 594ms/step - loss: 0.7916 - accuracy: 0.7111 - val_loss: 0.6188 - val_accuracy: 0.8055 - lr: 2.5000e-04\n", "Epoch 21/40\n", "84/84 [==============================] - 50s 594ms/step - loss: 0.7444 - accuracy: 0.7412 - val_loss: 0.6344 - val_accuracy: 0.8024 - lr: 2.5000e-04\n", "Epoch 22/40\n", "84/84 [==============================] - 51s 603ms/step - loss: 0.7086 - accuracy: 0.7585 - val_loss: 0.5986 - val_accuracy: 0.7964 - lr: 2.5000e-04\n", "Epoch 23/40\n", "84/84 [==============================] - 50s 588ms/step - loss: 0.7257 - accuracy: 0.7366 - val_loss: 0.6382 - val_accuracy: 0.7933 - lr: 2.5000e-04\n", "Epoch 24/40\n", "84/84 [==============================] - ETA: 0s - loss: 0.6831 - accuracy: 0.7622 \n", "Epoch 24: ReduceLROnPlateau reducing learning rate to 0.0001250000059371814.\n", "84/84 [==============================] - 50s 597ms/step - loss: 0.6831 - accuracy: 0.7622 - val_loss: 0.6136 - val_accuracy: 0.7872 - lr: 2.5000e-04\n", "Epoch 25/40\n", "84/84 [==============================] - 50s 595ms/step - loss: 0.6406 - accuracy: 0.7833 - val_loss: 0.6038 - val_accuracy: 0.8055 - lr: 1.2500e-04\n", "Epoch 26/40\n", "84/84 [==============================] - 51s 602ms/step - loss: 0.6423 - accuracy: 0.7713 - val_loss: 0.5395 - val_accuracy: 0.8298 - lr: 1.2500e-04\n", "Epoch 27/40\n", "84/84 [==============================] - 50s 591ms/step - loss: 0.5904 - accuracy: 0.7878 - val_loss: 0.5348 - val_accuracy: 0.8176 - lr: 1.2500e-04\n", "Epoch 28/40\n", "84/84 [==============================] - 49s 583ms/step - loss: 0.5719 - accuracy: 0.8044 - val_loss: 0.4902 - val_accuracy: 0.8419 - lr: 1.2500e-04\n", "Epoch 29/40\n", "84/84 [==============================] - 49s 585ms/step - loss: 0.5679 - accuracy: 0.8141 - val_loss: 0.5753 - val_accuracy: 0.8085 - lr: 1.2500e-04\n", "Epoch 30/40\n", "84/84 [==============================] - ETA: 0s - loss: 0.5491 - accuracy: 0.8126 \n", "Epoch 30: ReduceLROnPlateau reducing learning rate to 6.25000029685907e-05.\n", "84/84 [==============================] - 50s 589ms/step - loss: 0.5491 - accuracy: 0.8126 - val_loss: 0.5053 - val_accuracy: 0.8359 - lr: 1.2500e-04\n", "Epoch 31/40\n", "84/84 [==============================] - 50s 597ms/step - loss: 0.5126 - accuracy: 0.8262 - val_loss: 0.4782 - val_accuracy: 0.8389 - lr: 6.2500e-05\n", "Epoch 32/40\n", "84/84 [==============================] - 50s 592ms/step - loss: 0.5480 - accuracy: 0.8149 - val_loss: 0.4499 - val_accuracy: 0.8541 - lr: 6.2500e-05\n", "Epoch 33/40\n", "84/84 [==============================] - 50s 593ms/step - loss: 0.5307 - accuracy: 0.7991 - val_loss: 0.5397 - val_accuracy: 0.8207 - lr: 6.2500e-05\n", "Epoch 34/40\n", "84/84 [==============================] - ETA: 0s - loss: 0.5124 - accuracy: 0.8149 \n", "Epoch 34: ReduceLROnPlateau reducing learning rate to 3.125000148429535e-05.\n", "84/84 [==============================] - 50s 598ms/step - loss: 0.5124 - accuracy: 0.8149 - val_loss: 0.4712 - val_accuracy: 0.8419 - lr: 6.2500e-05\n", "Epoch 35/40\n", "84/84 [==============================] - 50s 600ms/step - loss: 0.4940 - accuracy: 0.8284 - val_loss: 0.4426 - val_accuracy: 0.8571 - lr: 3.1250e-05\n", "Epoch 36/40\n", "84/84 [==============================] - 51s 602ms/step - loss: 0.5285 - accuracy: 0.8149 - val_loss: 0.4847 - val_accuracy: 0.8450 - lr: 3.1250e-05\n", "Epoch 37/40\n", "84/84 [==============================] - ETA: 0s - loss: 0.5365 - accuracy: 0.8172 \n", "Epoch 37: ReduceLROnPlateau reducing learning rate to 1.5625000742147677e-05.\n", "84/84 [==============================] - 51s 603ms/step - loss: 0.5365 - accuracy: 0.8172 - val_loss: 0.4507 - val_accuracy: 0.8571 - lr: 3.1250e-05\n", "Epoch 38/40\n", "84/84 [==============================] - 52s 615ms/step - loss: 0.5049 - accuracy: 0.8269 - val_loss: 0.4969 - val_accuracy: 0.8480 - lr: 1.5625e-05\n", "Epoch 39/40\n", "84/84 [==============================] - ETA: 0s - loss: 0.5025 - accuracy: 0.8262 \n", "Epoch 39: ReduceLROnPlateau reducing learning rate to 7.812500371073838e-06.\n", "84/84 [==============================] - 50s 597ms/step - loss: 0.5025 - accuracy: 0.8262 - val_loss: 0.4632 - val_accuracy: 0.8480 - lr: 1.5625e-05\n", "Epoch 40/40\n", "84/84 [==============================] - 51s 605ms/step - loss: 0.4991 - accuracy: 0.8149 - val_loss: 0.4591 - val_accuracy: 0.8541 - lr: 7.8125e-06\n" ] } ], "source": [ "history = model.fit(train_data,validation_data=val_data,epochs=40,callbacks=[early_stop, checkpoint, lr_schedule])" ] }, { "cell_type": "code", "execution_count": 46, "id": "406a1b56-2b61-47c1-8758-65e469b39292", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "21/21 [==============================] - 9s 398ms/step - loss: 0.4426 - accuracy: 0.8571\n", "0.44259175658226013 0.8571428656578064\n" ] } ], "source": [ "model = tf.keras.models.load_model(\"best_date_cnn.keras\") \n", "test_loss, test_acc = model.evaluate(val_data)\n", "print(test_loss, test_acc)" ] }, { "cell_type": "code", "execution_count": 51, "id": "2f2d6cd7-4712-486e-ab69-df05d0223d24", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "21/21 [==============================] - 10s 442ms/step\n" ] } ], "source": [ "# Predictions of Validation\n", "y_pred = model.predict(val_data)\n", "y_pred_classes = np.argmax(y_pred, axis=1)\n", "y_true = val_data.classes" ] }, { "cell_type": "markdown", "id": "e5887a79-5db6-4cce-93b9-9489e8aa9793", "metadata": {}, "source": [ "## Methods\n", "- Input: RGB images with 224×224×3 \n", "- Architecture: 5 conv layers, 3 max pooling, Dense(256), Dropout(0.5) \n", "- Optimizer: Adam (lr=0.001) \n", "- Loss: Sparse Categorical Crossentropy \n", "- Regularization: EarlyStopping, ReduceLROnPlateau, ModelCheckpoint" ] }, { "cell_type": "code", "execution_count": 52, "id": "87ac7af6-1018-4214-bf97-244bdaf4ea49", "metadata": {}, "outputs": [], "source": [ "# Class names\n", "class_names = list(val_data.class_indices.keys())" ] }, { "cell_type": "code", "execution_count": 54, "id": "653b03ae-8df1-4631-b7ce-7623da1755ac", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Accuracy & Loss\n", "plt.figure(figsize=(14, 5))\n", "\n", "plt.subplot(1, 2, 1)\n", "plt.plot(history.history['accuracy'], label='Train Accuracy')\n", "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n", "plt.title('Model Accuracy')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Accuracy')\n", "plt.legend()\n", "\n", "plt.subplot(1, 2, 2)\n", "plt.plot(history.history['loss'], label='Train Loss')\n", "plt.plot(history.history['val_loss'], label='Validation Loss')\n", "plt.title('Model Loss')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Loss')\n", "plt.legend()\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 55, "id": "2aab0256-3952-4044-be4d-26dcb6b8a9a4", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Confusion matrix\n", "cm = confusion_matrix(y_true, y_pred_classes)\n", "\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.show()" ] }, { "cell_type": "code", "execution_count": 56, "id": "c5daad98-6047-42b3-b1be-034b5bf57b54", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Classification Report\n", " precision recall f1-score support\n", "\n", " Ajwa 1.00 1.00 1.00 35\n", " Galaxy 0.89 0.87 0.88 38\n", " Medjool 0.68 0.93 0.78 27\n", " Meneifi 0.73 0.65 0.69 46\n", " Nabtat Ali 0.86 0.91 0.89 35\n", " Rutab 1.00 1.00 1.00 29\n", " Shaishe 0.92 1.00 0.96 34\n", " Sokari 0.89 0.81 0.85 52\n", " Sugaey 0.76 0.67 0.71 33\n", "\n", " accuracy 0.86 329\n", " macro avg 0.86 0.87 0.86 329\n", "weighted avg 0.86 0.86 0.86 329\n", "\n" ] } ], "source": [ "# Classification Report\n", "print(\"Classification Report\")\n", "print(classification_report(y_true, y_pred_classes, target_names=class_names))" ] }, { "cell_type": "markdown", "id": "8d1fdd7a-aa05-462d-998b-b669eaba7d7d", "metadata": {}, "source": [] }, { "cell_type": "markdown", "id": "86e498df-2a13-4038-97fc-d08609b2d3f5", "metadata": {}, "source": [ "## Results\n", "Accuracy improves steadily and loss decreases consistently. Train and validation curves are similar → very little overfitting. The confusion matrix shows strong classification performance with only minor mistakes between similar date types." ] }, { "cell_type": "markdown", "id": "81bd601b-6668-45aa-8df5-fe598c47d9fb", "metadata": {}, "source": [ "## Reflection\n", "The model fully meets the project requirements. Even without transfer learning, it achieves high performance. For future improvements, one could:\n", "- apply more augmentation,\n", "- add more convolution blocks,\n", "- try transfer learning to possibly increase accuracy further." ] }, { "cell_type": "markdown", "id": "244ac620-21ad-44cf-8be3-66bb76089066", "metadata": {}, "source": [ "# Transferlearning ResNet50" ] }, { "cell_type": "markdown", "id": "13be2972-bb2e-47a4-a436-7cb2a6ea8e4e", "metadata": {}, "source": [ "## Introduction\n", "The goal of this project was to classify different types of date fruits using deep learning. Since the dataset is relatively small and the visual differences between the fruit types are subtle, Transfer Learning was applied using a pre-trained ResNet50 model." ] }, { "cell_type": "code", "execution_count": 10, "id": "90a1e22d-837a-4e79-871d-6bdb2ad56b33", "metadata": {}, "outputs": [], "source": [ "from tensorflow.keras.applications import ResNet50\n", "from tensorflow.keras.applications.resnet50 import preprocess_input" ] }, { "cell_type": "code", "execution_count": 11, "id": "e16ac8de-8ca1-4a9f-a39c-f119ac397125", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Found 1329 images belonging to 9 classes.\n", "Found 329 images belonging to 9 classes.\n" ] } ], "source": [ "img_size = (224, 224)\n", "batch_size = 16\n", "\n", "train_datagen = ImageDataGenerator(\n", " preprocessing_function=preprocess_input, \n", " validation_split=0.2,\n", " rotation_range=25,\n", " width_shift_range=0.15,\n", " height_shift_range=0.15,\n", " zoom_range=0.15,\n", " shear_range=0.15,\n", " horizontal_flip=True,\n", " fill_mode='nearest'\n", ")\n", "\n", "val_datagen = ImageDataGenerator(\n", " preprocessing_function=preprocess_input, \n", " validation_split=0.2\n", ")\n", "\n", "train_data = train_datagen.flow_from_directory(\n", " data_dir,\n", " target_size=img_size,\n", " batch_size=batch_size,\n", " subset='training',\n", " class_mode='sparse',\n", " shuffle=True\n", ")\n", "\n", "val_data = val_datagen.flow_from_directory(\n", " data_dir,\n", " target_size=img_size,\n", " batch_size=batch_size,\n", " subset='validation',\n", " class_mode='sparse',\n", " shuffle=False\n", ")" ] }, { "cell_type": "code", "execution_count": 12, "id": "4bb4cb46-0ca1-4157-b39c-50f8ec07c340", "metadata": {}, "outputs": [], "source": [ "callbacks = [\n", " ReduceLROnPlateau(monitor='val_loss', factor=0.5, patience=3, min_lr=1e-7),\n", " ModelCheckpoint(filepath='best_date_resnet50.keras', monitor='val_loss', save_best_only=True)\n", "]" ] }, { "cell_type": "code", "execution_count": 13, "id": "bdf2305d-c21c-4f91-944d-d6ea0e3bc50d", "metadata": {}, "outputs": [], "source": [ "base_model = ResNet50(\n", " weights='imagenet',\n", " include_top=False,\n", " input_shape=(224, 224, 3)\n", ")\n", "\n", "base_model.trainable = False\n", "\n", "model = Sequential()\n", "model.add(base_model)\n", "model.add(GlobalAveragePooling2D())\n", "model.add(Dense(256, activation='relu'))\n", "model.add(Dropout(0.5))\n", "model.add(Dense(num_classes, activation='softmax'))\n", "\n", "model.compile(\n", " optimizer=Adam(learning_rate=1e-3),\n", " loss='sparse_categorical_crossentropy',\n", " metrics=['accuracy']\n", ")" ] }, { "cell_type": "code", "execution_count": 14, "id": "d35f6f9c-0f8c-439f-a1bf-500a808b6c0a", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Epoch 1/40\n", " 1/84 [..............................] - ETA: 6:36 - loss: 3.1581 - accuracy: 0.1875" ] }, { "name": "stderr", "output_type": "stream", "text": [ "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n", "'+ptx85' is not a recognized feature for this target (ignoring feature)\n" ] }, { "name": "stdout", "output_type": "stream", "text": [ "84/84 [==============================] - 94s 1s/step - loss: 1.2588 - accuracy: 0.5922 - val_loss: 0.5364 - val_accuracy: 0.8328 - lr: 0.0010\n", "Epoch 2/40\n", "84/84 [==============================] - 75s 894ms/step - loss: 0.5597 - accuracy: 0.8126 - val_loss: 0.3333 - val_accuracy: 0.8967 - lr: 0.0010\n", "Epoch 3/40\n", "84/84 [==============================] - 131s 2s/step - loss: 0.4601 - accuracy: 0.8435 - val_loss: 0.3832 - val_accuracy: 0.8723 - lr: 0.0010\n", "Epoch 4/40\n", "84/84 [==============================] - 92s 1s/step - loss: 0.3903 - accuracy: 0.8728 - val_loss: 0.4710 - val_accuracy: 0.8419 - lr: 0.0010\n", "Epoch 5/40\n", "84/84 [==============================] - 84s 990ms/step - loss: 0.3583 - accuracy: 0.8834 - val_loss: 0.2741 - val_accuracy: 0.9088 - lr: 0.0010\n", "Epoch 6/40\n", "84/84 [==============================] - 79s 914ms/step - loss: 0.3664 - accuracy: 0.8713 - val_loss: 0.2776 - val_accuracy: 0.9027 - lr: 0.0010\n", "Epoch 7/40\n", "84/84 [==============================] - 70s 833ms/step - loss: 0.3328 - accuracy: 0.8864 - val_loss: 0.2647 - val_accuracy: 0.9301 - lr: 0.0010\n", "Epoch 8/40\n", "84/84 [==============================] - 82s 975ms/step - loss: 0.3023 - accuracy: 0.8999 - val_loss: 0.3231 - val_accuracy: 0.8967 - lr: 0.0010\n", "Epoch 9/40\n", "84/84 [==============================] - 96s 1s/step - loss: 0.2507 - accuracy: 0.9090 - val_loss: 0.3333 - val_accuracy: 0.8906 - lr: 0.0010\n", "Epoch 10/40\n", "84/84 [==============================] - 113s 1s/step - loss: 0.2460 - accuracy: 0.9097 - val_loss: 0.2882 - val_accuracy: 0.9088 - lr: 0.0010\n", "Epoch 11/40\n", "84/84 [==============================] - 112s 1s/step - loss: 0.1984 - accuracy: 0.9300 - val_loss: 0.3603 - val_accuracy: 0.8784 - lr: 5.0000e-04\n", "Epoch 12/40\n", "84/84 [==============================] - 98s 1s/step - loss: 0.1881 - accuracy: 0.9353 - val_loss: 0.2112 - val_accuracy: 0.9331 - lr: 5.0000e-04\n", "Epoch 13/40\n", "84/84 [==============================] - 98s 1s/step - loss: 0.1692 - accuracy: 0.9398 - val_loss: 0.1728 - val_accuracy: 0.9514 - lr: 5.0000e-04\n", "Epoch 14/40\n", "84/84 [==============================] - 101s 1s/step - loss: 0.1867 - accuracy: 0.9443 - val_loss: 0.1629 - val_accuracy: 0.9422 - lr: 5.0000e-04\n", "Epoch 15/40\n", "84/84 [==============================] - 86s 1s/step - loss: 0.1866 - accuracy: 0.9391 - val_loss: 0.3099 - val_accuracy: 0.9027 - lr: 5.0000e-04\n", "Epoch 16/40\n", "84/84 [==============================] - 122s 1s/step - loss: 0.1511 - accuracy: 0.9496 - val_loss: 0.1605 - val_accuracy: 0.9362 - lr: 5.0000e-04\n", "Epoch 17/40\n", "84/84 [==============================] - 105s 1s/step - loss: 0.1668 - accuracy: 0.9451 - val_loss: 0.2381 - val_accuracy: 0.9179 - lr: 5.0000e-04\n", "Epoch 18/40\n", "84/84 [==============================] - 50s 588ms/step - loss: 0.1535 - accuracy: 0.9533 - val_loss: 0.2707 - val_accuracy: 0.9058 - lr: 5.0000e-04\n", "Epoch 19/40\n", "84/84 [==============================] - 52s 615ms/step - loss: 0.1484 - accuracy: 0.9473 - val_loss: 0.1681 - val_accuracy: 0.9331 - lr: 5.0000e-04\n", "Epoch 20/40\n", "84/84 [==============================] - 47s 563ms/step - loss: 0.1287 - accuracy: 0.9609 - val_loss: 0.1711 - val_accuracy: 0.9574 - lr: 2.5000e-04\n", "Epoch 21/40\n", "84/84 [==============================] - 47s 554ms/step - loss: 0.1324 - accuracy: 0.9541 - val_loss: 0.1947 - val_accuracy: 0.9240 - lr: 2.5000e-04\n", "Epoch 22/40\n", "84/84 [==============================] - 51s 605ms/step - loss: 0.1166 - accuracy: 0.9624 - val_loss: 0.2045 - val_accuracy: 0.9210 - lr: 2.5000e-04\n", "Epoch 23/40\n", "84/84 [==============================] - 48s 567ms/step - loss: 0.1256 - accuracy: 0.9646 - val_loss: 0.1555 - val_accuracy: 0.9544 - lr: 1.2500e-04\n", "Epoch 24/40\n", "84/84 [==============================] - 53s 637ms/step - loss: 0.0896 - accuracy: 0.9691 - val_loss: 0.1501 - val_accuracy: 0.9605 - lr: 1.2500e-04\n", "Epoch 25/40\n", "84/84 [==============================] - 49s 577ms/step - loss: 0.0898 - accuracy: 0.9699 - val_loss: 0.1466 - val_accuracy: 0.9605 - lr: 1.2500e-04\n", "Epoch 26/40\n", "84/84 [==============================] - 47s 561ms/step - loss: 0.1097 - accuracy: 0.9676 - val_loss: 0.1871 - val_accuracy: 0.9422 - lr: 1.2500e-04\n", "Epoch 27/40\n", "84/84 [==============================] - 50s 601ms/step - loss: 0.1006 - accuracy: 0.9669 - val_loss: 0.1740 - val_accuracy: 0.9422 - lr: 1.2500e-04\n", "Epoch 28/40\n", "84/84 [==============================] - 48s 568ms/step - loss: 0.0942 - accuracy: 0.9661 - val_loss: 0.1692 - val_accuracy: 0.9483 - lr: 1.2500e-04\n", "Epoch 29/40\n", "84/84 [==============================] - 50s 596ms/step - loss: 0.0910 - accuracy: 0.9722 - val_loss: 0.1616 - val_accuracy: 0.9483 - lr: 6.2500e-05\n", "Epoch 30/40\n", "84/84 [==============================] - 48s 566ms/step - loss: 0.0857 - accuracy: 0.9699 - val_loss: 0.1901 - val_accuracy: 0.9301 - lr: 6.2500e-05\n", "Epoch 31/40\n", "84/84 [==============================] - 50s 591ms/step - loss: 0.0863 - accuracy: 0.9699 - val_loss: 0.1712 - val_accuracy: 0.9453 - lr: 6.2500e-05\n", "Epoch 32/40\n", "84/84 [==============================] - 52s 615ms/step - loss: 0.0854 - accuracy: 0.9729 - val_loss: 0.1730 - val_accuracy: 0.9483 - lr: 3.1250e-05\n", "Epoch 33/40\n", "84/84 [==============================] - 48s 568ms/step - loss: 0.0910 - accuracy: 0.9714 - val_loss: 0.1830 - val_accuracy: 0.9453 - lr: 3.1250e-05\n", "Epoch 34/40\n", "84/84 [==============================] - 51s 608ms/step - loss: 0.0992 - accuracy: 0.9699 - val_loss: 0.1781 - val_accuracy: 0.9392 - lr: 3.1250e-05\n", "Epoch 35/40\n", "84/84 [==============================] - 48s 574ms/step - loss: 0.0827 - accuracy: 0.9767 - val_loss: 0.1819 - val_accuracy: 0.9362 - lr: 1.5625e-05\n", "Epoch 36/40\n", "84/84 [==============================] - 50s 598ms/step - loss: 0.0848 - accuracy: 0.9691 - val_loss: 0.1740 - val_accuracy: 0.9453 - lr: 1.5625e-05\n", "Epoch 37/40\n", "84/84 [==============================] - 48s 571ms/step - loss: 0.0970 - accuracy: 0.9722 - val_loss: 0.1832 - val_accuracy: 0.9422 - lr: 1.5625e-05\n", "Epoch 38/40\n", "84/84 [==============================] - 48s 570ms/step - loss: 0.0705 - accuracy: 0.9789 - val_loss: 0.1852 - val_accuracy: 0.9422 - lr: 7.8125e-06\n", "Epoch 39/40\n", "84/84 [==============================] - 50s 588ms/step - loss: 0.0877 - accuracy: 0.9722 - val_loss: 0.1824 - val_accuracy: 0.9453 - lr: 7.8125e-06\n", "Epoch 40/40\n", "84/84 [==============================] - 47s 562ms/step - loss: 0.0951 - accuracy: 0.9699 - val_loss: 0.1842 - val_accuracy: 0.9422 - lr: 7.8125e-06\n" ] } ], "source": [ "history = model.fit(train_data,validation_data=val_data,epochs=40,callbacks=callbacks,verbose=1 )" ] }, { "cell_type": "code", "execution_count": 15, "id": "90c0c65e-f234-46c0-a521-fdfb31a99e39", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "21/21 [==============================] - 8s 342ms/step - loss: 0.1842 - accuracy: 0.9422\n", "Validation Loss: 0.1842\n", "Validation Accuracy: 0.9422\n" ] } ], "source": [ "# test modell\n", "val_loss, val_acc = model.evaluate(val_data)\n", "print(f\"Validation Loss: {val_loss:.4f}\")\n", "print(f\"Validation Accuracy: {val_acc:.4f}\")" ] }, { "cell_type": "code", "execution_count": 19, "id": "83dc7344-135e-4372-a75c-f5566a587393", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# Accuracy and Loss-Plots from history\n", "plt.figure(figsize=(14, 5))\n", "\n", "plt.subplot(1, 2, 1)\n", "plt.plot(history.history['accuracy'], label='Train Accuracy')\n", "plt.plot(history.history['val_accuracy'], label='Validation Accuracy')\n", "plt.title('Model Accuracy')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Accuracy')\n", "plt.legend()\n", "\n", "plt.subplot(1, 2, 2)\n", "plt.plot(history.history['loss'], label='Train Loss')\n", "plt.plot(history.history['val_loss'], label='Validation Loss')\n", "plt.title('Model Loss')\n", "plt.xlabel('Epoch')\n", "plt.ylabel('Loss')\n", "plt.legend()\n", "\n", "plt.show()" ] }, { "cell_type": "code", "execution_count": 20, "id": "e33c751a-05dc-4025-a3e9-a680b15c3a08", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "['Ajwa', 'Galaxy', 'Medjool', 'Meneifi', 'Nabtat Ali', 'Rutab', 'Shaishe', 'Sokari', 'Sugaey']\n" ] } ], "source": [ "# fetching class names\n", "class_indices = val_data.class_indices \n", "idx_to_class = {v: k for k, v in class_indices.items()}\n", "class_names = [idx_to_class[i] for i in range(len(idx_to_class))]\n", "print(class_names)" ] }, { "cell_type": "code", "execution_count": 21, "id": "f78f98f7-5274-425a-84ee-1f8f8782da1b", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "21/21 [==============================] - 12s 521ms/step\n" ] } ], "source": [ "# calculate probabilities\n", "val_data.reset() \n", "\n", "# probabilities\n", "y_prob = model.predict(val_data)\n", "\n", "# predictet classes\n", "y_pred = np.argmax(y_prob, axis=1)\n", "\n", "# Labels from Generator\n", "y_true = val_data.classes" ] }, { "cell_type": "code", "execution_count": 22, "id": "3b9fcaea-e514-403a-880c-76ec90bab8b2", "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ "Classification Report:\n", " precision recall f1-score support\n", "\n", " Ajwa 1.00 1.00 1.00 35\n", " Galaxy 0.97 0.97 0.97 38\n", " Medjool 0.71 0.93 0.81 27\n", " Meneifi 1.00 0.72 0.84 46\n", " Nabtat Ali 0.95 1.00 0.97 35\n", " Rutab 0.94 1.00 0.97 29\n", " Shaishe 1.00 1.00 1.00 34\n", " Sokari 0.96 1.00 0.98 52\n", " Sugaey 0.94 0.91 0.92 33\n", "\n", " accuracy 0.94 329\n", " macro avg 0.94 0.95 0.94 329\n", "weighted avg 0.95 0.94 0.94 329\n", "\n" ] } ], "source": [ "# Confusion Matrix & Classification Report\n", "cm = confusion_matrix(y_true, y_pred)\n", "print(\"Classification Report:\")\n", "print(classification_report(y_true, y_pred, target_names=class_names))" ] }, { "cell_type": "code", "execution_count": 23, "id": "f4a70859-45f3-40e9-860f-8d41fb5236c0", "metadata": {}, "outputs": [ { "data": { "image/png": "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", "text/plain": [ "
" ] }, "metadata": {}, "output_type": "display_data" } ], "source": [ "# plot confusion matrix\n", "fig, ax = plt.subplots(figsize=(6, 6))\n", "im = ax.imshow(cm, interpolation='nearest')\n", "ax.figure.colorbar(im, ax=ax)\n", "\n", "ax.set(\n", " xticks=np.arange(len(class_names)),\n", " yticks=np.arange(len(class_names)),\n", " xticklabels=class_names,\n", " yticklabels=class_names,\n", " ylabel='True label',\n", " xlabel='Predicted label',\n", " title='Confusion Matrix'\n", ")\n", "\n", "plt.setp(ax.get_xticklabels(), rotation=45, ha=\"right\",\n", " rotation_mode=\"anchor\")\n", "\n", "thresh = cm.max() / 2.\n", "for i in range(cm.shape[0]):\n", " for j in range(cm.shape[1]):\n", " ax.text(j, i, cm[i, j],\n", " ha=\"center\", va=\"center\",\n", " color=\"white\" if cm[i, j] > thresh else \"black\")\n", "\n", "plt.tight_layout()\n", "plt.show()" ] }, { "cell_type": "markdown", "id": "efc657d7-14d7-4613-9a7e-e3a431bf744e", "metadata": {}, "source": [ "## Analysis\n", "The dataset contains multiple date fruit classes such as Ajwa, Galaxy, Medjool, Meneifi, Nabtat Ali, Rutab, Shaishe, Sokari, and Sugaey.\n", "\n", "Images were resized to 224×224 pixels to match the input size of ResNet50.\n", "\n", "A 20% validation split was used.\n", "\n", "Some classes look visually similar, which can cause misclassifications.\n", "\n", "No missing or incorrect labels were detected." ] }, { "cell_type": "markdown", "id": "027443b2-3856-4278-8596-b0dea92981a4", "metadata": {}, "source": [ "## Methods\n", "\n", "Data Preprocessing\n", "\n", "Images were processed using preprocess_input (required for ResNet50).\n", "\n", "Augmentation applied: rotation (25°), width/height shift (0.15), zoom (0.15), shear (0.15), and horizontal flip.\n", "\n", "Batch size: 16.\n", "\n", "Model Architecture\n", "\n", "ResNet50 (weights='imagenet', include_top=False, trainable=False)\n", "→ GlobalAveragePooling2D\n", "→ Dense(256, activation='relu')\n", "→ Dropout(0.5)\n", "→ Dense(num_classes, activation='softmax')" ] }, { "cell_type": "markdown", "id": "cd3f4e90-1cc3-492c-86d3-ed282b00f487", "metadata": {}, "source": [ "## Results\n", "\n", "### Overall Model Performance\n", "\n", "| Metric | Result |\n", "|--------|--------|\n", "| **Validation Accuracy** | **94.22%** |\n", "| **Validation Loss** | **0.1842** |\n", "| **Macro Avg F1-score** | **0.94** |\n", "\n", "\n", "### Performance Per Class\n", "\n", "| Class | Precision | Recall | F1-score |\n", "|------------|-----------|--------|----------|\n", "| **Ajwa** | 1.00 | 1.00 | 1.00 |\n", "| **Galaxy** | 0.97 | 0.97 | 0.97 |\n", "| **Medjool** | 0.71 | 0.93 | 0.81 |\n", "| **Meneifi** | 1.00 | 0.72 | 0.84 |\n", "| **Nabtat Ali** | 0.95 | 1.00 | 0.97 |\n", "| **Rutab** | 0.94 | 1.00 | 0.97 |\n", "| **Shaishe** | 1.00 | 1.00 | 1.00 |\n", "| **Sokari** | 0.96 | 1.00 | 0.98 |\n", "| **Sugaey** | 0.94 | 0.91 | 0.92 |" ] }, { "cell_type": "markdown", "id": "6f310650-92fc-4f8e-bed7-b25df0608fc0", "metadata": {}, "source": [ "### CNN vs Transfer Learning (ResNet50)\n", "\n", "| Model | Validation Accuracy | Validation Loss | F1-score |\n", "|-------|---------------------|-----------------|----------|\n", "| **Custom CNN (ohne TL)** | z. B. 90% | ~0.30 | ~0.88 |\n", "| **ResNet50 (mit TL)** | **94.22%** | **0.1842** | **0.94** |" ] }, { "cell_type": "markdown", "id": "03892f50-e8e2-422e-a1af-709da253435e", "metadata": {}, "source": [ "## Reflection\n", "\n", "Transfer Learning with ResNet50 performed significantly better than the custom CNN model. The model learned quickly and generalized well to unseen images. There was no relevant overfitting.\n", "\n", "The final accuracy of 94.22% demonstrates that Transfer Learning was the correct approach for this task." ] } ], "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": "3.12.3" } }, "nbformat": 4, "nbformat_minor": 5 }