{"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"code","source":"# This Python 3 environment comes with many helpful analytics libraries installed\n# It is defined by the kaggle/python Docker image: https://github.com/kaggle/docker-python\n# For example, here's several helpful packages to load\n\nimport numpy as np # linear algebra\nimport pandas as pd # data processing, CSV file I/O (e.g. pd.read_csv)\nimport tensorflow as tf\nfrom kaggle_datasets import KaggleDatasets\n# Input data files are available in the read-only \"../input/\" directory\n# For example, running this (by clicking run or pressing Shift+Enter) will list all files under the input directory\n\nimport os\nfor dirname, _, filenames in os.walk('/kaggle/input'):\n    for filename in filenames:\n        print(os.path.join(dirname, filename))\n\nprint(\"Tensorflow version \" + tf.__version__)\n# You can write up to 20GB to the current directory (/kaggle/working/) that gets preserved as output when you create a version using \"Save & Run All\" \n# You can also write temporary files to /kaggle/temp/, but they won't be saved outside of the current session","metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","execution":{"iopub.status.busy":"2023-11-13T04:31:36.071560Z","iopub.execute_input":"2023-11-13T04:31:36.071904Z","iopub.status.idle":"2023-11-13T04:31:50.268426Z","shell.execute_reply.started":"2023-11-13T04:31:36.071861Z","shell.execute_reply":"2023-11-13T04:31:50.267689Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Detect Accelerator","metadata":{}},{"cell_type":"code","source":"# Detect hardware, return appropriate distribution strategy\ntry:\n    tpu = tf.distribute.cluster_resolver.TPUClusterResolver()  \n    # TPU detection. No parameters necessary if TPU_NAME environment variable is set. \n    # On Kaggle this is always the case.\n    print('Running on TPU ', tpu.master())\nexcept ValueError:\n    tpu = None\n\nif tpu:\n    tf.config.experimental_connect_to_cluster(tpu)\n    tf.tpu.experimental.initialize_tpu_system(tpu)\n    strategy = tf.distribute.experimental.TPUStrategy(tpu)\nelse:\n    strategy = tf.distribute.get_strategy() \n    # default distribution strategy in Tensorflow. Works on CPU and single GPU.\n\nprint(\"REPLICAS: \", strategy.num_replicas_in_sync)","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:31:50.270051Z","iopub.execute_input":"2023-11-13T04:31:50.270526Z","iopub.status.idle":"2023-11-13T04:31:58.713121Z","shell.execute_reply.started":"2023-11-13T04:31:50.270486Z","shell.execute_reply":"2023-11-13T04:31:58.712226Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Set Some Parameters","metadata":{}},{"cell_type":"code","source":"IMAGE_SIZE = [192, 192] # at this size, a GPU will run out of memory. Use the TPU\nEPOCHS = 5\nBATCH_SIZE = 16 * strategy.num_replicas_in_sync\n\nNUM_TRAINING_IMAGES = 12753\nNUM_TEST_IMAGES = 7382\nSTEPS_PER_EPOCH = NUM_TRAINING_IMAGES // BATCH_SIZE","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:31:58.714202Z","iopub.execute_input":"2023-11-13T04:31:58.714454Z","iopub.status.idle":"2023-11-13T04:31:58.718602Z","shell.execute_reply.started":"2023-11-13T04:31:58.714427Z","shell.execute_reply":"2023-11-13T04:31:58.717941Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Data Preprocessing and Data Loading","metadata":{}},{"cell_type":"code","source":"GCS_DS_PATH = KaggleDatasets().get_gcs_path() # you can list the bucket with \"!gsutil ls $GCS_DS_PATH\"\n\ndef decode_image(image_data):\n    image = tf.image.decode_jpeg(image_data, channels=3)\n    image = tf.cast(image, tf.float32) / 255.0  # convert image to floats in [0, 1] range\n    image = tf.reshape(image, [*IMAGE_SIZE, 3]) # explicit size needed for TPU\n    return image\n\ndef read_labeled_tfrecord(example):\n    LABELED_TFREC_FORMAT = {\n        \"image\": tf.io.FixedLenFeature([], tf.string), # tf.string means bytestring\n        \"class\": tf.io.FixedLenFeature([], tf.int64),  # shape [] means single element\n    }\n    example = tf.io.parse_single_example(example, LABELED_TFREC_FORMAT)\n    image = decode_image(example['image'])\n    label = tf.cast(example['class'], tf.int32)\n    return image, label # returns a dataset of (image, label) pairs\n\ndef read_unlabeled_tfrecord(example):\n    UNLABELED_TFREC_FORMAT = {\n        \"image\": tf.io.FixedLenFeature([], tf.string), # tf.string means bytestring\n        \"id\": tf.io.FixedLenFeature([], tf.string),  # shape [] means single element\n        # class is missing, this competitions's challenge is to predict flower classes for the test dataset\n    }\n    example = tf.io.parse_single_example(example, UNLABELED_TFREC_FORMAT)\n    image = decode_image(example['image'])\n    idnum = example['id']\n    return image, idnum # returns a dataset of image(s)\n\ndef load_dataset(filenames, labeled=True, ordered=False):\n    # Read from TFRecords. For optimal performance, reading from multiple files at once and\n    # disregarding data order. Order does not matter since we will be shuffling the data anyway.\n\n    ignore_order = tf.data.Options()\n    if not ordered:\n        ignore_order.experimental_deterministic = False # disable order, increase speed\n\n    dataset = tf.data.TFRecordDataset(filenames) # automatically interleaves reads from multiple files\n    dataset = dataset.with_options(ignore_order) # uses data as soon as it streams in, rather than in its original order\n    dataset = dataset.map(read_labeled_tfrecord if labeled else read_unlabeled_tfrecord)\n    # returns a dataset of (image, label) pairs if labeled=True or (image, id) pairs if labeled=False\n    return dataset\n\ndef get_training_dataset():\n    dataset = load_dataset(tf.io.gfile.glob(GCS_DS_PATH + '/tfrecords-jpeg-192x192/train/*.tfrec'), labeled=True)\n    dataset = dataset.repeat() # the training dataset must repeat for several epochs\n    dataset = dataset.shuffle(2048)\n    dataset = dataset.batch(BATCH_SIZE)\n    return dataset\n\ndef get_validation_dataset():\n    dataset = load_dataset(tf.io.gfile.glob(GCS_DS_PATH + '/tfrecords-jpeg-192x192/val/*.tfrec'), labeled=True, ordered=False)\n    dataset = dataset.batch(BATCH_SIZE)\n    dataset = dataset.cache()\n    return dataset\n\ndef get_test_dataset(ordered=False):\n    dataset = load_dataset(tf.io.gfile.glob(GCS_DS_PATH + '/tfrecords-jpeg-192x192/test/*.tfrec'), labeled=False, ordered=ordered)\n    dataset = dataset.batch(BATCH_SIZE)\n    return dataset\n\ntraining_dataset = get_training_dataset()\nvalidation_dataset = get_validation_dataset()","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:31:58.719512Z","iopub.execute_input":"2023-11-13T04:31:58.719788Z","iopub.status.idle":"2023-11-13T04:31:58.898644Z","shell.execute_reply.started":"2023-11-13T04:31:58.719761Z","shell.execute_reply":"2023-11-13T04:31:58.897783Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Build a model","metadata":{}},{"cell_type":"code","source":"learning_rate = 0.02\nwith strategy.scope():    \n    pretrained_model = tf.keras.applications.VGG16(weights='imagenet', include_top=False ,input_shape=[*IMAGE_SIZE, 3])\n    pretrained_model.trainable = False # tramsfer learning\n    \n    model = tf.keras.Sequential([\n        pretrained_model,\n        tf.keras.layers.GlobalAveragePooling2D(),\n        tf.keras.layers.Dense(104, activation='softmax')\n    ])\n        \nmodel.compile(\n    optimizer=tf.keras.optimizers.Adam(learning_rate=learning_rate),\n    loss = 'sparse_categorical_crossentropy',\n    metrics=['sparse_categorical_accuracy']\n)\n\nhistorical = model.fit(training_dataset, \n          steps_per_epoch=STEPS_PER_EPOCH, \n          epochs=EPOCHS, \n          validation_data=validation_dataset)","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:31:58.900426Z","iopub.execute_input":"2023-11-13T04:31:58.900689Z","iopub.status.idle":"2023-11-13T04:32:50.598098Z","shell.execute_reply.started":"2023-11-13T04:31:58.900661Z","shell.execute_reply":"2023-11-13T04:32:50.596829Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Validation Dataset Accuracy","metadata":{}},{"cell_type":"code","source":"# Evaluate model on the validation dataset and print accuracy\nvalidation_accuracy = model.evaluate(validation_dataset)[1]\nprint(\"Validation Accuracy:\", validation_accuracy)","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:32:50.600031Z","iopub.execute_input":"2023-11-13T04:32:50.600337Z","iopub.status.idle":"2023-11-13T04:32:52.496986Z","shell.execute_reply.started":"2023-11-13T04:32:50.600305Z","shell.execute_reply":"2023-11-13T04:32:52.495912Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Predictions on the Test Dataset","metadata":{}},{"cell_type":"code","source":"test_ds = get_test_dataset(ordered=True) # since we are splitting the dataset and iterating separately on images and ids, order matters.\n\nprint('Computing predictions...')\ntest_images_ds = test_ds.map(lambda image, idnum: image)\nprobabilities = model.predict(test_images_ds)\npredictions = np.argmax(probabilities, axis=-1)\nprint(predictions)\n\nprint('Generating submission.csv file...')\ntest_ids_ds = test_ds.map(lambda image, idnum: idnum).unbatch()\ntest_ids = next(iter(test_ids_ds.batch(NUM_TEST_IMAGES))).numpy().astype('U') # all in one batch\nnp.savetxt('submission.csv', np.rec.fromarrays([test_ids, predictions]), fmt=['%s', '%d'], delimiter=',', header='id,label', comments='')","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:32:52.498247Z","iopub.execute_input":"2023-11-13T04:32:52.498580Z","iopub.status.idle":"2023-11-13T04:33:06.842875Z","shell.execute_reply.started":"2023-11-13T04:32:52.498544Z","shell.execute_reply":"2023-11-13T04:33:06.841642Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Experimenting with different learning rates and Visualizing the findings","metadata":{}},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\n# Conduct experiments by varying a hyperparameter (e.g., learning rate)\nlearning_rates = [0.001, 0.01, 0.1]\nvalidation_accuracies = []\n\nfor lr in learning_rates:\n    print(f'\\nExperimenting with Learning Rate: {lr}\\n')\n\n    # Re-create the model with the new learning rate\n    with strategy.scope():    \n        pretrained_model = tf.keras.applications.VGG16(weights='imagenet', include_top=False, input_shape=[*IMAGE_SIZE, 3])\n        pretrained_model.trainable = False\n        \n        model = tf.keras.Sequential([\n            pretrained_model,\n            tf.keras.layers.GlobalAveragePooling2D(),\n            tf.keras.layers.Dense(104, activation='softmax')\n        ])\n    \n    model.compile(\n        optimizer=tf.keras.optimizers.Adam(learning_rate=lr),\n        loss='sparse_categorical_crossentropy',\n        metrics=['sparse_categorical_accuracy']\n    )\n    \n    # Train the model\n    history = model.fit(training_dataset, \n                        steps_per_epoch=STEPS_PER_EPOCH, \n                        epochs=EPOCHS, \n                        validation_data=validation_dataset,\n                        verbose=0)  # Set verbose to 0 to suppress training logs\n\n    # Store validation accuracy for each experiment\n    validation_accuracy = history.history['val_sparse_categorical_accuracy'][-1]\n    validation_accuracies.append(validation_accuracy)\n\n    # Plot the training and validation accuracy for each experiment\n    plt.plot(history.history['sparse_categorical_accuracy'], label='Training Accuracy (LR=' + str(lr) + ')')\n    plt.plot(history.history['val_sparse_categorical_accuracy'], label='Validation Accuracy (LR=' + str(lr) + ')')\n\n# Display the plot\nplt.title('Training and Validation Accuracy for Different Learning Rates')\nplt.xlabel('Epochs')\nplt.ylabel('Accuracy')\nplt.legend()\nplt.show()\n\n# Print the validation accuracies for each experiment\nfor lr, acc in zip(learning_rates, validation_accuracies):\n    print(f'Validation Accuracy (LR={lr}): {acc}')","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:33:06.844014Z","iopub.execute_input":"2023-11-13T04:33:06.844284Z","iopub.status.idle":"2023-11-13T04:35:22.574122Z","shell.execute_reply.started":"2023-11-13T04:33:06.844256Z","shell.execute_reply":"2023-11-13T04:35:22.573253Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import matplotlib.pyplot as plt\n\n# Conduct experiments by varying a hyperparameter (e.g., learning rate)\nlearning_rates = [0.001, 0.005, 0.01, 0.02, 0.05, 0.1]\nvalidation_accuracies = []\n\nfor lr in learning_rates:\n    print(f'\\nExperimenting with Learning Rate: {lr}\\n')\n\n    # Re-create the model with the new learning rate\n    with strategy.scope():    \n        pretrained_model = tf.keras.applications.VGG16(weights='imagenet', include_top=False, input_shape=[*IMAGE_SIZE, 3])\n        pretrained_model.trainable = False\n        \n        model = tf.keras.Sequential([\n            pretrained_model,\n            tf.keras.layers.GlobalAveragePooling2D(),\n            tf.keras.layers.Dense(104, activation='softmax')\n        ])\n    \n    model.compile(\n        optimizer=tf.keras.optimizers.Adam(learning_rate=lr),\n        loss='sparse_categorical_crossentropy',\n        metrics=['sparse_categorical_accuracy']\n    )\n    \n    # Train the model\n    history = model.fit(training_dataset, \n                        steps_per_epoch=STEPS_PER_EPOCH, \n                        epochs=EPOCHS, \n                        validation_data=validation_dataset,\n                        verbose=0)  # Set verbose to 0 to suppress training logs\n\n    # Store validation accuracy for each experiment\n    validation_accuracy = history.history['val_sparse_categorical_accuracy'][-1]\n    validation_accuracies.append(validation_accuracy)\n\n    # Plot the training and validation accuracy for each experiment\n    plt.plot(history.history['sparse_categorical_accuracy'], label='Training Accuracy (LR=' + str(lr) + ')')\n    plt.plot(history.history['val_sparse_categorical_accuracy'], label='Validation Accuracy (LR=' + str(lr) + ')')\n\n# Display the plot\nplt.title('Training and Validation Accuracy for Different Learning Rates')\nplt.xlabel('Epochs')\nplt.ylabel('Accuracy')\nplt.legend()\nplt.show()\n\n# Print the validation accuracies for each experiment\nfor lr, acc in zip(learning_rates, validation_accuracies):\n    print(f'Validation Accuracy (LR={lr}): {acc}')","metadata":{"execution":{"iopub.status.busy":"2023-11-13T04:35:22.575188Z","iopub.execute_input":"2023-11-13T04:35:22.575473Z","iopub.status.idle":"2023-11-13T04:39:54.864712Z","shell.execute_reply.started":"2023-11-13T04:35:22.575444Z","shell.execute_reply":"2023-11-13T04:39:54.863569Z"},"trusted":true},"execution_count":null,"outputs":[]}]}