{"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":"markdown","source":"Learning how to use [KerasTuner](https://keras.io/keras_tuner/) to build a better Tensorflow model.\n\n## TODO\n\n- The tuning should not happen when submitting to the leaderboard.  Save the best parameters for LB.\n\n## Versions\n\n- V4: Tuning filters and units\n```python\n    x = layers.Dense(\n        units=hp.Int('num_dense_units', min_value=16, max_value=64, step=8),\n        activation='relu'\n    )(x)\n```    \n- V3: Documentation\n- V2: Documentation\n- V1: Only tuning Dropout()\n\n## References\n\n- [Keras + KerasTuner best practices](https://www.kaggle.com/fchollet/keras-kerastuner-best-practices)\n- [MoA: Keras + KerasTuner best practices](https://www.kaggle.com/fchollet/moa-keras-kerastuner-best-practices)\n- [Deep Learning with Python, Second Edition](https://www.manning.com/books/deep-learning-with-python-second-edition) - Chapter 13\n- [Jane Street: Neural Network Starter](https://www.kaggle.com/gogo827jz/jane-street-neural-network-starter)\n- [NN Model tuning with Keras Tuner](https://www.kaggle.com/sirishapb/nn-model-tuning-with-keras-tuner)\n","metadata":{"execution":{"iopub.status.busy":"2021-10-06T14:51:45.864214Z","iopub.execute_input":"2021-10-06T14:51:45.864969Z","iopub.status.idle":"2021-10-06T14:51:45.888709Z","shell.execute_reply.started":"2021-10-06T14:51:45.864882Z","shell.execute_reply":"2021-10-06T14:51:45.887418Z"}}},{"cell_type":"markdown","source":"# Load Libraries","metadata":{}},{"cell_type":"code","source":"import os\nimport glob\n\nimport pandas as pd\nimport numpy as np\nfrom pathlib import Path\n\nimport random\nfrom tqdm.notebook import tqdm\nimport pydicom # Handle MRI images\n\nimport cv2  # OpenCV - https://docs.opencv.org/master/d6/d00/tutorial_py_root.html\n\nfrom sklearn.model_selection import train_test_split\nfrom sklearn.metrics import roc_auc_score\n\n\nimport tensorflow as tf\nfrom tensorflow import keras\nfrom tensorflow.keras.utils import to_categorical\nfrom tensorflow.keras import layers\n","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:50:13.708065Z","iopub.execute_input":"2021-10-07T00:50:13.708637Z","iopub.status.idle":"2021-10-07T00:50:18.596966Z","shell.execute_reply.started":"2021-10-07T00:50:13.708536Z","shell.execute_reply":"2021-10-07T00:50:18.596164Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Load Datasets","metadata":{}},{"cell_type":"code","source":"data_dir = Path('../input/rsna-miccai-brain-tumor-radiogenomic-classification/')\n\nmri_types = [\"FLAIR\", \"T1w\", \"T2w\", \"T1wCE\"]\nexcluded_images = [109, 123, 709] # Bad images\n\ntrain_df = pd.read_csv(data_dir / \"train_labels.csv\")\ntest_df = pd.read_csv(data_dir / \"sample_submission.csv\")\nsample_submission = pd.read_csv(data_dir / \"sample_submission.csv\")\n\ntrain_df = train_df[~train_df.BraTS21ID.isin(excluded_images)]\n","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:50:18.59867Z","iopub.execute_input":"2021-10-07T00:50:18.598925Z","iopub.status.idle":"2021-10-07T00:50:18.633896Z","shell.execute_reply.started":"2021-10-07T00:50:18.598892Z","shell.execute_reply":"2021-10-07T00:50:18.633204Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Utility Functions","metadata":{}},{"cell_type":"code","source":"def load_dicom(path, size = 512):\n    ''' \n    Reads a DICOM image, standardizes so that the pixel values are between 0 and 1, then rescales to 0 and 255\n    \n    Not super sure if this kind of scaling is appropriate, but everyone seems to do it. \n    '''\n    dicom = pydicom.read_file(path)\n    data = dicom.pixel_array\n    # transform data into black and white scale / grayscale\n#     data = data - np.min(data)\n    if np.max(data) != 0:\n        data = data / np.max(data)\n    data = (data * 255).astype(np.uint8)\n    return cv2.resize(data, (size, size))\n\ndef get_all_image_paths(brats21id, image_type, folder='train'): \n    '''\n    Returns an arry of all the images of a particular type for a particular patient ID\n    '''\n    assert(image_type in mri_types)\n    \n    patient_path = os.path.join(\n        \"../input/rsna-miccai-brain-tumor-radiogenomic-classification/%s/\" % folder, \n        str(brats21id).zfill(5),\n    )\n\n    paths = sorted(\n        glob.glob(os.path.join(patient_path, image_type, \"*\")), \n        key=lambda x: int(x[:-4].split(\"-\")[-1]),\n    )\n    \n    num_images = len(paths)\n    \n    start = int(num_images * 0.25)\n    end = int(num_images * 0.75)\n\n    interval = 3\n    \n    if num_images < 10: \n        interval = 1\n    \n    return np.array(paths[start:end:interval])\n\ndef get_all_images(brats21id, image_type, folder='train', size=512):\n    return [load_dicom(path, size) for path in get_all_image_paths(brats21id, image_type, folder)]\n\ndef get_all_data_for_train(image_type, image_size=413):\n    global train_df\n    \n    X = []\n    y = []\n    train_ids = []\n\n    for i in tqdm(train_df.index):\n        x = train_df.loc[i]\n        images = get_all_images(int(x['BraTS21ID']), image_type, 'train', image_size)\n        label = x['MGMT_value']\n\n        X += images\n        y += [label] * len(images)\n        train_ids += [int(x['BraTS21ID'])] * len(images)\n        assert(len(X) == len(y))\n    return np.array(X), np.array(y), np.array(train_ids)\n\ndef get_all_data_for_test(image_type, image_size=512):\n    global test_df\n    \n    X = []\n    test_ids = []\n\n    for i in tqdm(test_df.index):\n        x = test_df.loc[i]\n        images = get_all_images(int(x['BraTS21ID']), image_type, 'test', image_size)\n        X += images\n        test_ids += [int(x['BraTS21ID'])] * len(images)\n\n    return np.array(X), np.array(test_ids)","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:50:18.635439Z","iopub.execute_input":"2021-10-07T00:50:18.635758Z","iopub.status.idle":"2021-10-07T00:50:18.654004Z","shell.execute_reply.started":"2021-10-07T00:50:18.635723Z","shell.execute_reply":"2021-10-07T00:50:18.652707Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Load all Images\n```\nX - contains all the images for each patient \ntrainidt - trainidt is a mask vector into X, y for training.  There's a patient id/BraTS21ID corresponding to each image (e.g. (0, 0, 0, 0, 2,2, 3,3,3,3,3,...) )\ntestidt - testidt is a mask vector into X_test for testing\n```","metadata":{}},{"cell_type":"code","source":"X, y, trainidt = get_all_data_for_train('T1wCE', image_size=32)\nX_test, testidt = get_all_data_for_test('T1wCE', image_size=32)","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:50:18.655702Z","iopub.execute_input":"2021-10-07T00:50:18.655989Z","iopub.status.idle":"2021-10-07T00:52:40.52701Z","shell.execute_reply.started":"2021-10-07T00:50:18.655951Z","shell.execute_reply":"2021-10-07T00:52:40.526167Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Train/Validation Split","metadata":{}},{"cell_type":"code","source":"X_train, X_valid, y_train, y_valid, trainidt_train, trainidt_valid = train_test_split(X, y, trainidt, random_state=12)","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:52:40.529013Z","iopub.execute_input":"2021-10-07T00:52:40.529351Z","iopub.status.idle":"2021-10-07T00:52:40.544173Z","shell.execute_reply.started":"2021-10-07T00:52:40.529313Z","shell.execute_reply":"2021-10-07T00:52:40.543379Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Adding a Dimension","metadata":{}},{"cell_type":"code","source":"X_train = tf.expand_dims(X_train, axis=-1)\nX_valid = tf.expand_dims(X_valid, axis=-1)\nX_train.shape","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:52:40.545759Z","iopub.execute_input":"2021-10-07T00:52:40.546168Z","iopub.status.idle":"2021-10-07T00:52:42.49229Z","shell.execute_reply.started":"2021-10-07T00:52:40.546128Z","shell.execute_reply":"2021-10-07T00:52:42.491587Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## One-hot encode labels","metadata":{}},{"cell_type":"code","source":"y_train = to_categorical(y_train)\ny_valid = to_categorical(y_valid)","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:52:42.493727Z","iopub.execute_input":"2021-10-07T00:52:42.493997Z","iopub.status.idle":"2021-10-07T00:52:42.498805Z","shell.execute_reply.started":"2021-10-07T00:52:42.493963Z","shell.execute_reply":"2021-10-07T00:52:42.498118Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Tunable Model","metadata":{}},{"cell_type":"code","source":"import keras_tuner as kt\n\n\ndef make_model(hp):\n    inputs = keras.Input(shape=X_train.shape[1:])\n    \n    x = keras.layers.experimental.preprocessing.Rescaling(1.0 / 255)(inputs)\n\n#     num_block = hp.Int('num_block', min_value=2, max_value=5, step=1)\n#     num_filters = hp.Int('num_filters', min_value=32, max_value=128, step=32)\n    \n#     x = keras.layers.Conv2D(64, kernel_size=(4, 4), activation=\"relu\", name=\"Conv_1\")(x)\n    x = keras.layers.Conv2D(filters=hp.Int('units_Conv_1_' + str(0),\n                                            min_value=64,\n                                            max_value=256,\n                                            step=32),\n                            kernel_size=(4, 4),\n                            activation=\"relu\", \n                            name=\"Conv_1\")(x)\n\n    x = keras.layers.MaxPool2D(pool_size=(2, 2))(x)\n\n#     x = keras.layers.Conv2D(32, kernel_size=(2, 2), activation=\"relu\", name=\"Conv_2\")(x)\n    x = keras.layers.Conv2D(filters=hp.Int('units_conv2_' + str(1),\n                                            min_value=16,\n                                            max_value=128,\n                                            step=16),\n                            kernel_size=(2, 2),\n                            activation=\"relu\",\n                            name=\"Conv_2\")(x)\n\n    x = keras.layers.MaxPool2D(pool_size=(1, 1))(x)\n    \n#     for i in range(num_block):\n#         x = keras.layers.Conv2D(num_filters, \n#                                 kernel_size=(4, 4),\n#                                 activation=\"relu\",\n#                                 )(x)\n    \n#         x = keras.layers.MaxPool2D(pool_size=(2, 2))(x)\n\n#     x = keras.layers.Conv2D(32, kernel_size=(2, 2), activation=\"relu\", name=\"Conv_2\")(x)\n#     x = keras.layers.MaxPool2D(pool_size=(1, 1))(x)\n\n#     h = keras.layers.Dropout(0.1)(h)\n    x = layers.Dropout(\n        hp.Float('dense_dropout', min_value=0., max_value=0.7)\n    )(x)\n    x = keras.layers.Flatten()(x)\n#     reduction_type = hp.Choice('reduction_type', ['flatten', 'avg'])\n#     if reduction_type == 'flatten':\n#         x = layers.Flatten()(x)\n#     else:\n#         x = layers.GlobalAveragePooling2D()(x)\n        \n#     x = keras.layers.Dense(32, activation=\"relu\")(x)\n    x = layers.Dense(\n        units=hp.Int('num_dense_units', min_value=16, max_value=64, step=8),\n        activation='relu'\n    )(x)\n\n    outputs = keras.layers.Dense(2, activation=\"softmax\")(x)\n\n    model = keras.Model(inputs, outputs)\n\n    roc_auc = tf.keras.metrics.AUC(name='roc_auc', curve='ROC')\n\n    model.compile(\n        loss=\"categorical_crossentropy\", optimizer=\"adam\", metrics=[roc_auc]\n    )\n    model.summary()\n    return model","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:58:41.356071Z","iopub.execute_input":"2021-10-07T00:58:41.356402Z","iopub.status.idle":"2021-10-07T00:58:41.368722Z","shell.execute_reply.started":"2021-10-07T00:58:41.356369Z","shell.execute_reply":"2021-10-07T00:58:41.368026Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Hyperparameter Search","metadata":{}},{"cell_type":"code","source":"tuner = kt.tuners.BayesianOptimization(\n    make_model,\n    objective='val_loss',\n    max_trials=13,  # Set to 5 to run quicker, but need 100+ for good results\n    overwrite=True)\n\ncallbacks=[keras.callbacks.EarlyStopping(monitor='val_roc_acc', mode='max', patience=3, baseline=0.9)]\n\ntuner.search(X_train, y_train, validation_split=0.2 ,callbacks=callbacks, verbose=1, epochs=200)","metadata":{"execution":{"iopub.status.busy":"2021-10-07T00:58:42.379589Z","iopub.execute_input":"2021-10-07T00:58:42.380124Z","iopub.status.idle":"2021-10-07T01:01:54.946145Z","shell.execute_reply.started":"2021-10-07T00:58:42.38009Z","shell.execute_reply":"2021-10-07T01:01:54.945395Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Find the best epoch value","metadata":{}},{"cell_type":"code","source":"best_hp = tuner.get_best_hyperparameters()[0]\nbest_model = make_model(best_hp)\nhistory = best_model.fit(X_train, y_train,validation_split=0.2, epochs=150)","metadata":{"execution":{"iopub.status.busy":"2021-10-06T22:14:41.505736Z","iopub.execute_input":"2021-10-06T22:14:41.506592Z","iopub.status.idle":"2021-10-06T22:15:55.901976Z","shell.execute_reply.started":"2021-10-06T22:14:41.50655Z","shell.execute_reply":"2021-10-06T22:15:55.901289Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Predictions on Validation Set","metadata":{}},{"cell_type":"code","source":"y_pred = best_model.predict(X_valid)\n\npred = np.argmax(y_pred, axis=1)\n\nresult = pd.DataFrame(trainidt_valid)\nresult[1] = pred\n\nresult.columns = [\"BraTS21ID\", \"MGMT_value\"]\nresult2 = result.groupby(\"BraTS21ID\", as_index=False).mean()\n\nresult2 = result2.merge(train_df, on=\"BraTS21ID\")\nauc = roc_auc_score(\n    result2.MGMT_value_y,\n    result2.MGMT_value_x,\n)\nprint(f\"Validation AUC={auc}\")\n","metadata":{"execution":{"iopub.status.busy":"2021-10-06T22:15:55.903688Z","iopub.execute_input":"2021-10-06T22:15:55.903962Z","iopub.status.idle":"2021-10-06T22:15:56.146195Z","shell.execute_reply.started":"2021-10-06T22:15:55.903926Z","shell.execute_reply":"2021-10-06T22:15:56.1455Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Predictions on the Test Set","metadata":{}},{"cell_type":"code","source":"y_pred = best_model.predict(X_test)\n\npred = np.argmax(y_pred, axis=1) #\n\nresult = pd.DataFrame(testidt)\nresult[1] = pred\npred\n","metadata":{"execution":{"iopub.status.busy":"2021-10-06T22:15:56.147396Z","iopub.execute_input":"2021-10-06T22:15:56.147668Z","iopub.status.idle":"2021-10-06T22:15:56.326192Z","shell.execute_reply.started":"2021-10-06T22:15:56.147635Z","shell.execute_reply":"2021-10-06T22:15:56.32553Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# Submission File","metadata":{}},{"cell_type":"code","source":"result.columns=['BraTS21ID','MGMT_value']\n\nresult2 = result.groupby('BraTS21ID',as_index=False).mean()\nresult2['BraTS21ID'] = sample_submission['BraTS21ID']\n\nresult2['MGMT_value'] = result2['MGMT_value'] # No rounding\nresult2.to_csv('submission.csv',index=False)\nresult2\n","metadata":{"execution":{"iopub.status.busy":"2021-10-06T22:15:56.327381Z","iopub.execute_input":"2021-10-06T22:15:56.32766Z","iopub.status.idle":"2021-10-06T22:15:56.356394Z","shell.execute_reply.started":"2021-10-06T22:15:56.327623Z","shell.execute_reply":"2021-10-06T22:15:56.355699Z"},"trusted":true},"execution_count":null,"outputs":[]}]}