Deep Learning Projects/transfer-learning-with-resnetv250-efficientnetb0.ipynb

{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"name":"python","version":"3.10.12","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat_minor":4,"nbformat":4,"cells":[{"source":"<a href=\"https://www.kaggle.com/code/kelixirr/transfer-learning-with-resnetv250-efficientnetb0?scriptVersionId=142375324\" target=\"_blank\"><img align=\"left\" alt=\"Kaggle\" title=\"Open in Kaggle\" src=\"https://kaggle.com/static/images/open-in-kaggle.svg\"></a>","metadata":{},"cell_type":"markdown"},{"cell_type":"code","source":"import zipfile\n\n!wget https://storage.googleapis.com/ztm_tf_course/food_vision/10_food_classes_10_percent.zip \n    \nzip_ref = zipfile.ZipFile(\"10_food_classes_10_percent.zip\", \"r\")\nzip_ref.extractall()\nzip_ref.close() ","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:20:48.254447Z","iopub.execute_input":"2023-09-08T20:20:48.254871Z","iopub.status.idle":"2023-09-08T20:20:57.300043Z","shell.execute_reply.started":"2023-09-08T20:20:48.254836Z","shell.execute_reply":"2023-09-08T20:20:57.298897Z"},"trusted":true},"execution_count":1,"outputs":[{"name":"stdout","text":"--2023-09-08 20:20:49--  https://storage.googleapis.com/ztm_tf_course/food_vision/10_food_classes_10_percent.zip\nResolving storage.googleapis.com (storage.googleapis.com)... 64.233.187.207, 64.233.188.207, 173.194.174.207, ...\nConnecting to storage.googleapis.com (storage.googleapis.com)|64.233.187.207|:443... connected.\nHTTP request sent, awaiting response... 200 OK\nLength: 168546183 (161M) [application/zip]\nSaving to: ‘10_food_classes_10_percent.zip’\n\n10_food_classes_10_ 100%[===================>] 160.74M  29.9MB/s    in 6.3s    \n\n2023-09-08 20:20:55 (25.5 MB/s) - ‘10_food_classes_10_percent.zip’ saved [168546183/168546183]\n\n","output_type":"stream"}]},{"cell_type":"code","source":"import os\n\nfor dirpath, dirnames, filenames in os.walk(\"10_food_classes_10_percent\"):\n    print(f\"There are {len(dirnames)} directories and {len(filenames)} images in '{dirpath}'.\")","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:21:02.339304Z","iopub.execute_input":"2023-09-08T20:21:02.339725Z","iopub.status.idle":"2023-09-08T20:21:02.353801Z","shell.execute_reply.started":"2023-09-08T20:21:02.339693Z","shell.execute_reply":"2023-09-08T20:21:02.352605Z"},"trusted":true},"execution_count":2,"outputs":[{"name":"stdout","text":"There are 2 directories and 0 images in '10_food_classes_10_percent'.\nThere are 10 directories and 0 images in '10_food_classes_10_percent/test'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/hamburger'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/steak'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/pizza'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/chicken_curry'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/grilled_salmon'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/chicken_wings'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/sushi'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/fried_rice'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/ice_cream'.\nThere are 0 directories and 250 images in '10_food_classes_10_percent/test/ramen'.\nThere are 10 directories and 0 images in '10_food_classes_10_percent/train'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/hamburger'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/steak'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/pizza'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/chicken_curry'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/grilled_salmon'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/chicken_wings'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/sushi'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/fried_rice'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/ice_cream'.\nThere are 0 directories and 75 images in '10_food_classes_10_percent/train/ramen'.\n","output_type":"stream"}]},{"cell_type":"markdown","source":"Preparing Data","metadata":{}},{"cell_type":"code","source":"from tensorflow.keras.preprocessing.image import ImageDataGenerator\n\nIMAGE_SHAPE = (224, 224)\nBATCH_SIZE = 32\n\ntrain_dir = \"10_food_classes_10_percent/train/\" \ntest_dir = \"10_food_classes_10_percent/test/\" \n\ntrain_datagen = ImageDataGenerator(rescale = 1/255.) \ntest_datagen = ImageDataGenerator(rescale = 1/255.) \n\ntrain_data = train_datagen.flow_from_directory(train_dir,\n                                              target_size = IMAGE_SHAPE,\n                                              batch_size = BATCH_SIZE,\n                                              class_mode = \"categorical\")\n\ntest_data = test_datagen.flow_from_directory(test_dir,\n                                            target_size = IMAGE_SHAPE,\n                                            batch_size = BATCH_SIZE,\\\n                                            class_mode = \"categorical\")","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:27:37.329599Z","iopub.execute_input":"2023-09-08T20:27:37.330022Z","iopub.status.idle":"2023-09-08T20:27:37.544252Z","shell.execute_reply.started":"2023-09-08T20:27:37.329986Z","shell.execute_reply":"2023-09-08T20:27:37.54322Z"},"trusted":true},"execution_count":3,"outputs":[{"name":"stdout","text":"Found 750 images belonging to 10 classes.\nFound 2500 images belonging to 10 classes.\n","output_type":"stream"}]},{"cell_type":"code","source":"import tensorflow as tf\nimport tensorflow_hub as hub \nfrom tensorflow.keras import layers","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:35:11.76492Z","iopub.execute_input":"2023-09-08T20:35:11.765359Z","iopub.status.idle":"2023-09-08T20:35:12.193342Z","shell.execute_reply.started":"2023-09-08T20:35:11.765323Z","shell.execute_reply":"2023-09-08T20:35:12.192062Z"},"trusted":true},"execution_count":5,"outputs":[]},{"cell_type":"code","source":"# tensorboard callback\nimport datetime\n\ndef create_tensorboard_callback(dir_name, experiment_name):\n    log_dir = dir_name + \"/\" + experiment_name + \"/\" + datetime.datetime.now().strftime(\"%Y%m%d-%H%M%S\")\n    tensorboard_callback = tf.keras.callbacks.TensorBoard(log_dir = log_dir)\n    \n    print(f\"Saving TensorBoard log files to: {log_dir}\")\n    return tensorboard_callback","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:38:14.52584Z","iopub.execute_input":"2023-09-08T20:38:14.526395Z","iopub.status.idle":"2023-09-08T20:38:14.533497Z","shell.execute_reply.started":"2023-09-08T20:38:14.526339Z","shell.execute_reply":"2023-09-08T20:38:14.532477Z"},"trusted":true},"execution_count":6,"outputs":[]},{"cell_type":"code","source":"# Resnet 50 V2 feature vector\nresnet_url = \"https://tfhub.dev/google/imagenet/resnet_v2_50/feature_vector/4\"\n\n# Original: EfficientNetB0 feature vector (version 1)\nefficientnet_url = \"https://tfhub.dev/tensorflow/efficientnet/b0/feature-vector/1\" ","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:39:12.074928Z","iopub.execute_input":"2023-09-08T20:39:12.076109Z","iopub.status.idle":"2023-09-08T20:39:12.080998Z","shell.execute_reply.started":"2023-09-08T20:39:12.076071Z","shell.execute_reply":"2023-09-08T20:39:12.079721Z"},"trusted":true},"execution_count":7,"outputs":[]},{"cell_type":"code","source":"def create_model(model_url, num_classes = 10):\n    \n    # download the pre_trained model \n    \n    feature_extractor_layer = hub.KerasLayer(model_url,\n                                             trainable = False,\n                                             name = 'feature_extraction_layer',\n                                             input_shape = IMAGE_SHAPE+(3,))\n    \n    # our model \n    model = tf.keras.Sequential([   \n        feature_extractor_layer,\n        layers.Dense(num_classes, activation = 'softmax', name = 'output_layer')\n    ])\n    \n    return model","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:44:33.920522Z","iopub.execute_input":"2023-09-08T20:44:33.920943Z","iopub.status.idle":"2023-09-08T20:44:33.928133Z","shell.execute_reply.started":"2023-09-08T20:44:33.920901Z","shell.execute_reply":"2023-09-08T20:44:33.926896Z"},"trusted":true},"execution_count":8,"outputs":[]},{"cell_type":"markdown","source":"## ResNetV250","metadata":{}},{"cell_type":"code","source":"resnet_model = create_model(resnet_url, num_classes = train_data.num_classes)\n\nresnet_model.compile(loss = 'categorical_crossentropy',\n                     optimizer = tf.keras.optimizers.Adam(),\n                     metrics = ['accuracy'])\n\nresnet_history = resnet_model.fit(train_data,\n                                 epochs = 5,\n                                 steps_per_epoch = len(train_data),\n                                 validation_data = test_data,\n                                 validation_steps = len(test_data),\n                                 callbacks = [create_tensorboard_callback(dir_name=\"tensorflow_hub\", experiment_name=\"resnet50V2\")])","metadata":{"execution":{"iopub.status.busy":"2023-09-08T20:53:22.719979Z","iopub.execute_input":"2023-09-08T20:53:22.720373Z","iopub.status.idle":"2023-09-08T21:16:18.052155Z","shell.execute_reply.started":"2023-09-08T20:53:22.720344Z","shell.execute_reply":"2023-09-08T21:16:18.050925Z"},"trusted":true},"execution_count":13,"outputs":[{"name":"stdout","text":"Saving TensorBoard log files to: tensorflow_hub/resnet50V2/20230908-205325\nEpoch 1/5\n24/24 [==============================] - 273s 12s/step - loss: 1.8260 - accuracy: 0.3960 - val_loss: 1.1035 - val_accuracy: 0.6660\nEpoch 2/5\n24/24 [==============================] - 261s 11s/step - loss: 0.8601 - accuracy: 0.7533 - val_loss: 0.8271 - val_accuracy: 0.7332\nEpoch 3/5\n24/24 [==============================] - 258s 11s/step - loss: 0.6209 - accuracy: 0.8253 - val_loss: 0.7810 - val_accuracy: 0.7268\nEpoch 4/5\n24/24 [==============================] - 262s 11s/step - loss: 0.4684 - accuracy: 0.8720 - val_loss: 0.6887 - val_accuracy: 0.7720\nEpoch 5/5\n24/24 [==============================] - 254s 11s/step - loss: 0.3694 - accuracy: 0.9160 - val_loss: 0.6835 - val_accuracy: 0.7660\n","output_type":"stream"}]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\ndef plot_loss_curves(history):\n    \n    loss = history.history['loss']\n    val_loss = history.history['val_loss']\n    \n    accuracy = history.history['accuracy']\n    val_accuracy = history.history['val_accuracy']\n    \n    epochs = range(len(history.history['loss']))\n    \n    # plot loss \n    plt.plot(epochs, loss, label='training_loss')\n    plt.plot(epochs, val_loss, label = 'val_loss')\n    plt.title('Loss')\n    plt.xlabel('Epochs')\n    plt.legend()\n    \n    #plot accuracy \n    plt.plot(epochs, accuracy, label = 'training_accuracy')\n    plt.plot(epochs, val_accuracy, label = 'val_accuracy')\n    plt.title('Accuracy')\n    plt.xlabel('Epochs')\n    plt.legend()","metadata":{"execution":{"iopub.status.busy":"2023-09-08T21:25:47.643553Z","iopub.execute_input":"2023-09-08T21:25:47.643972Z","iopub.status.idle":"2023-09-08T21:25:47.65295Z","shell.execute_reply.started":"2023-09-08T21:25:47.643938Z","shell.execute_reply":"2023-09-08T21:25:47.651585Z"},"trusted":true},"execution_count":39,"outputs":[]},{"cell_type":"code","source":"plot_loss_curves(resnet_history)","metadata":{"execution":{"iopub.status.busy":"2023-09-08T21:26:01.260115Z","iopub.execute_input":"2023-09-08T21:26:01.260545Z","iopub.status.idle":"2023-09-08T21:26:01.63098Z","shell.execute_reply.started":"2023-09-08T21:26:01.260512Z","shell.execute_reply":"2023-09-08T21:26:01.629848Z"},"trusted":true},"execution_count":40,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"resnet_model.summary()","metadata":{"execution":{"iopub.status.busy":"2023-09-08T21:23:15.214169Z","iopub.execute_input":"2023-09-08T21:23:15.214578Z","iopub.status.idle":"2023-09-08T21:23:15.243897Z","shell.execute_reply.started":"2023-09-08T21:23:15.214546Z","shell.execute_reply":"2023-09-08T21:23:15.24279Z"},"trusted":true},"execution_count":27,"outputs":[{"name":"stdout","text":"Model: \"sequential_4\"\n_________________________________________________________________\n Layer (type)                Output Shape              Param #   \n=================================================================\n feature_extraction_layer (K  (None, 2048)             23564800  \n erasLayer)                                                      \n                                                                 \n output_layer (Dense)        (None, 10)                20490     \n                                                                 \n=================================================================\nTotal params: 23,585,290\nTrainable params: 20,490\nNon-trainable params: 23,564,800\n_________________________________________________________________\n","output_type":"stream"}]},{"cell_type":"markdown","source":"## EfficientNetB0","metadata":{}},{"cell_type":"code","source":"efficientnet_model = create_model(model_url = efficientnet_url,\n                                  num_classes = train_data.num_classes)\n\nefficientnet_model.compile(loss = 'categorical_crossentropy',\n                           optimizer = tf.keras.optimizers.Adam(),\n                           metrics = ['accuracy'])\n\nefficientnet_model.fit(train_data,\n                       epochs = 5,\n                       steps_per_epoch = len(train_data),\n                       validation_data = test_data,\n                       validation_steps = len(test_data),\n                       callbacks = [create_tensorboard_callback(dir_name=\"tensorflow_hub\", experiment_name=\"efficientnetB0\")])","metadata":{"execution":{"iopub.status.busy":"2023-09-08T21:27:55.739447Z","iopub.execute_input":"2023-09-08T21:27:55.739864Z","iopub.status.idle":"2023-09-08T21:42:26.791214Z","shell.execute_reply.started":"2023-09-08T21:27:55.739832Z","shell.execute_reply":"2023-09-08T21:42:26.789867Z"},"trusted":true},"execution_count":42,"outputs":[{"name":"stdout","text":"Saving TensorBoard log files to: tensorflow_hub/efficientnetB0/20230908-212810\nEpoch 1/5\n24/24 [==============================] - 142s 6s/step - loss: 1.9343 - accuracy: 0.3987 - val_loss: 1.3633 - val_accuracy: 0.7328\nEpoch 2/5\n24/24 [==============================] - 172s 7s/step - loss: 1.1065 - accuracy: 0.7747 - val_loss: 0.8989 - val_accuracy: 0.8152\nEpoch 3/5\n24/24 [==============================] - 172s 7s/step - loss: 0.7772 - accuracy: 0.8307 - val_loss: 0.7141 - val_accuracy: 0.8408\nEpoch 4/5\n24/24 [==============================] - 172s 7s/step - loss: 0.6208 - accuracy: 0.8733 - val_loss: 0.6236 - val_accuracy: 0.8604\nEpoch 5/5\n24/24 [==============================] - 129s 6s/step - loss: 0.5246 - accuracy: 0.8853 - val_loss: 0.5649 - val_accuracy: 0.8612\n","output_type":"stream"}]},{"cell_type":"code","source":"plot_loss_curves(efficientnet_model.history)","metadata":{"execution":{"iopub.status.busy":"2023-09-08T21:44:25.294368Z","iopub.execute_input":"2023-09-08T21:44:25.294788Z","iopub.status.idle":"2023-09-08T21:44:25.675631Z","shell.execute_reply.started":"2023-09-08T21:44:25.29476Z","shell.execute_reply":"2023-09-08T21:44:25.674479Z"},"trusted":true},"execution_count":44,"outputs":[{"output_type":"display_data","data":{"text/plain":"<Figure size 640x480 with 1 Axes>","image/png":"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"},"metadata":{}}]},{"cell_type":"code","source":"efficientnet_model.summary()","metadata":{"execution":{"iopub.status.busy":"2023-09-08T21:44:30.033892Z","iopub.execute_input":"2023-09-08T21:44:30.034312Z","iopub.status.idle":"2023-09-08T21:44:30.063301Z","shell.execute_reply.started":"2023-09-08T21:44:30.03428Z","shell.execute_reply":"2023-09-08T21:44:30.062125Z"},"trusted":true},"execution_count":45,"outputs":[{"name":"stdout","text":"Model: \"sequential_5\"\n_________________________________________________________________\n Layer (type)                Output Shape              Param #   \n=================================================================\n feature_extraction_layer (K  (None, 1280)             4049564   \n erasLayer)                                                      \n                                                                 \n output_layer (Dense)        (None, 10)                12810     \n                                                                 \n=================================================================\nTotal params: 4,062,374\nTrainable params: 12,810\nNon-trainable params: 4,049,564\n_________________________________________________________________\n","output_type":"stream"}]}]}