{"metadata":{"kaggle":{"accelerator":"gpu","dataSources":[{"sourceId":59093,"databundleVersionId":7469972,"sourceType":"competition"},{"sourceId":7526248,"sourceType":"datasetVersion","datasetId":4308295},{"sourceId":6127,"sourceType":"modelInstanceVersion","modelInstanceId":4598,"modelId":2797}],"isInternetEnabled":false,"language":"python","sourceType":"notebook","isGpuEnabled":true},"kernelspec":{"display_name":"Python 3","language":"python","name":"python3"},"language_info":{"name":"python","version":"3.10.14","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"},"papermill":{"default_parameters":{},"duration":3846.080383,"end_time":"2024-01-14T04:20:19.064569","environment_variables":{},"exception":null,"input_path":"__notebook__.ipynb","output_path":"__notebook__.ipynb","parameters":{},"start_time":"2024-01-14T03:16:12.984186","version":"2.4.0"},"widgets":{"application/vnd.jupyter.widget-state+json":{"state":{"08983a9c6aff42578980f4f7113c3ee2":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_4411aefc021d46d0ada7b645eb53ec48","placeholder":"โ€‹","style":"IPY_MODEL_09a10a8cf9334c51857397ed50398c8e","value":"Searching best thr : 100%"}},"09a10a8cf9334c51857397ed50398c8e":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"1f3989a0c01248328e16875075e9d1c4":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HBoxModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HBoxModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HBoxView","box_style":"","children":["IPY_MODEL_08983a9c6aff42578980f4f7113c3ee2","IPY_MODEL_22cfcc0a7cc6455fbf3bb7c788c8a4e1","IPY_MODEL_c8392e8075224e3b8a020a16c1a08447"],"layout":"IPY_MODEL_6cec9a2c2fac450d87248aed8dd62f86"}},"22cfcc0a7cc6455fbf3bb7c788c8a4e1":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"FloatProgressModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"FloatProgressModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"ProgressView","bar_style":"success","description":"","description_tooltip":null,"layout":"IPY_MODEL_dffe80502d954bdea0bbb6353dbf5515","max":20,"min":0,"orientation":"horizontal","style":"IPY_MODEL_7ce1b34a4f864a42a6619eec82311eb0","value":20}},"4411aefc021d46d0ada7b645eb53ec48":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"6cec9a2c2fac450d87248aed8dd62f86":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"7ce1b34a4f864a42a6619eec82311eb0":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"ProgressStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"ProgressStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","bar_color":null,"description_width":""}},"83fe40a0b8f047cc8602206909d42361":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}},"9384babdb7054d55aecdf3e989ddc926":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"DescriptionStyleModel","state":{"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"DescriptionStyleModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"StyleView","description_width":""}},"c8392e8075224e3b8a020a16c1a08447":{"model_module":"@jupyter-widgets/controls","model_module_version":"1.5.0","model_name":"HTMLModel","state":{"_dom_classes":[],"_model_module":"@jupyter-widgets/controls","_model_module_version":"1.5.0","_model_name":"HTMLModel","_view_count":null,"_view_module":"@jupyter-widgets/controls","_view_module_version":"1.5.0","_view_name":"HTMLView","description":"","description_tooltip":null,"layout":"IPY_MODEL_83fe40a0b8f047cc8602206909d42361","placeholder":"โ€‹","style":"IPY_MODEL_9384babdb7054d55aecdf3e989ddc926","value":" 20/20 [04:34<00:00, 12.66s/it]"}},"dffe80502d954bdea0bbb6353dbf5515":{"model_module":"@jupyter-widgets/base","model_module_version":"1.2.0","model_name":"LayoutModel","state":{"_model_module":"@jupyter-widgets/base","_model_module_version":"1.2.0","_model_name":"LayoutModel","_view_count":null,"_view_module":"@jupyter-widgets/base","_view_module_version":"1.2.0","_view_name":"LayoutView","align_content":null,"align_items":null,"align_self":null,"border":null,"bottom":null,"display":null,"flex":null,"flex_flow":null,"grid_area":null,"grid_auto_columns":null,"grid_auto_flow":null,"grid_auto_rows":null,"grid_column":null,"grid_gap":null,"grid_row":null,"grid_template_areas":null,"grid_template_columns":null,"grid_template_rows":null,"height":null,"justify_content":null,"justify_items":null,"left":null,"margin":null,"max_height":null,"max_width":null,"min_height":null,"min_width":null,"object_fit":null,"object_position":null,"order":null,"overflow":null,"overflow_x":null,"overflow_y":null,"padding":null,"right":null,"top":null,"visibility":null,"width":null}}},"version_major":2,"version_minor":0}}},"nbformat_minor":4,"nbformat":4,"cells":[{"cell_type":"markdown","source":"
\"Keras
\nThis starter notebook is provided by the Keras team.
","metadata":{"execution":{"iopub.execute_input":"2024-01-10T05:24:31.308329Z","iopub.status.busy":"2024-01-10T05:24:31.307595Z","iopub.status.idle":"2024-01-10T05:24:31.313088Z","shell.execute_reply":"2024-01-10T05:24:31.312113Z","shell.execute_reply.started":"2024-01-10T05:24:31.308287Z"},"papermill":{"duration":0.011755,"end_time":"2024-01-14T03:16:16.447481","exception":false,"start_time":"2024-01-14T03:16:16.435726","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"# HMS - Harmful Brain Activity Classification with [KerasCV](https://github.com/keras-team/keras-cv) and [Keras](https://github.com/keras-team/keras)\n\n> The objective of this competition is to classify seizures and other patterns of harmful brain activity in critically ill patients\n\nThis notebook guides you through the process of training and inferring a Deep Learning model, specifically EfficientNetV2, using KerasCV on the competition dataset. Specificaclly, this notebook uses spectrogram of the eeg data to classify the patterns.\n\nFun fact: This notebook is backend-agnostic, supporting TensorFlow, PyTorch, and JAX. Utilizing KerasCV and Keras allows us to choose our preferred backend. Explore more details on [Keras](https://keras.io/keras_core/announcement/).\n\nIn this notebook, you will learn:\n\n* Loading the data efficiently using [`tf.data`](https://www.tensorflow.org/guide/data).\n* Creating the model using KerasCV presets.\n* Training the model.\n* Inference and Submission on test data.\n\n**Note**: For a more in-depth understanding of KerasCV, refer to the [KerasCV guides](https://keras.io/guides/keras_cv/).","metadata":{}},{"cell_type":"markdown","source":"# ๐Ÿ›  | Install Libraries \n\nSince internet access is **disabled** during inference, we cannot install libraries in the usual `!pip install ` manner. Instead, we need to install libraries from local files. In the following cell, we will install libraries from our local files. The installation code stays very similar - we just use the `filepath` instead of the `filename` of the library. So now the code is `!pip install `. \n\n> The `filepath` of these local libraries look quite complicated, but don't be intimidated! Also `--no-deps` argument ensures that we are not installing any additional libraries.","metadata":{"papermill":{"duration":0.011416,"end_time":"2024-01-14T03:16:16.470167","exception":false,"start_time":"2024-01-14T03:16:16.458751","status":"completed"},"tags":[]}},{"cell_type":"code","source":"!pip install -q /kaggle/input/kerasv3-lib-ds/keras_cv-0.8.2-py3-none-any.whl --no-deps\n!pip install -q /kaggle/input/kerasv3-lib-ds/tensorflow-2.15.0.post1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl --no-deps\n!pip install -q /kaggle/input/kerasv3-lib-ds/keras-3.0.4-py3-none-any.whl --no-deps","metadata":{"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿ“š | Import Libraries ","metadata":{"papermill":{"duration":0.010878,"end_time":"2024-01-14T03:17:49.510159","exception":false,"start_time":"2024-01-14T03:17:49.499281","status":"completed"},"tags":[]}},{"cell_type":"code","source":"import os\nos.environ[\"KERAS_BACKEND\"] = \"jax\" # you can also use tensorflow or torch\n\nimport keras_cv\nimport keras\nfrom keras import ops\nimport tensorflow as tf\n\nimport cv2\nimport pandas as pd\nimport numpy as np\nfrom glob import glob\nfrom tqdm.notebook import tqdm\nimport joblib\n\nimport matplotlib.pyplot as plt ","metadata":{"papermill":{"duration":10.671979,"end_time":"2024-01-14T03:18:00.193134","exception":false,"start_time":"2024-01-14T03:17:49.521155","status":"completed"},"tags":[],"_kg_hide-output":true,"execution":{"iopub.status.busy":"2024-09-12T17:33:37.202659Z","iopub.execute_input":"2024-09-12T17:33:37.203324Z","iopub.status.idle":"2024-09-12T17:33:54.339996Z","shell.execute_reply.started":"2024-09-12T17:33:37.203288Z","shell.execute_reply":"2024-09-12T17:33:54.339169Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import os, gc\nos.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0,1\"\nimport tensorflow as tf\nimport pandas as pd, numpy as np\nimport matplotlib.pyplot as plt\nprint('TensorFlow version =',tf.__version__)\n\n# USE MULTIPLE GPUS\ngpus = tf.config.list_physical_devices('GPU')\nif len(gpus)<=1: \n strategy = tf.distribute.OneDeviceStrategy(device=\"/gpu:0\")\n print(f'Using {len(gpus)} GPU')\nelse: \n strategy = tf.distribute.MirroredStrategy()\n print(f'Using {len(gpus)} GPUs')\n\nVER = 5\n\n# IF THIS EQUALS NONE, THEN WE TRAIN NEW MODELS\n# IF THIS EQUALS DISK PATH, THEN WE LOAD PREVIOUSLY TRAINED MODELS\nLOAD_MODELS_FROM = '/kaggle/input/brain-efficientnet-models-v3-v4-v5/'\n\nUSE_KAGGLE_SPECTROGRAMS = True\nUSE_EEG_SPECTROGRAMS = True","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:33:57.125287Z","iopub.execute_input":"2024-09-12T17:33:57.126383Z","iopub.status.idle":"2024-09-12T17:33:57.317383Z","shell.execute_reply.started":"2024-09-12T17:33:57.126339Z","shell.execute_reply":"2024-09-12T17:33:57.316491Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Library Versions","metadata":{"papermill":{"duration":0.010958,"end_time":"2024-01-14T03:18:00.215704","exception":false,"start_time":"2024-01-14T03:18:00.204746","status":"completed"},"tags":[]}},{"cell_type":"code","source":"print(\"TensorFlow:\", tf.__version__)\nprint(\"Keras:\", keras.__version__)\nprint(\"KerasCV:\", keras_cv.__version__)","metadata":{"papermill":{"duration":0.019435,"end_time":"2024-01-14T03:18:00.246368","exception":false,"start_time":"2024-01-14T03:18:00.226933","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:33:58.733811Z","iopub.execute_input":"2024-09-12T17:33:58.734186Z","iopub.status.idle":"2024-09-12T17:33:58.739924Z","shell.execute_reply.started":"2024-09-12T17:33:58.73415Z","shell.execute_reply":"2024-09-12T17:33:58.739013Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# โš™๏ธ | Configuration","metadata":{"papermill":{"duration":0.010922,"end_time":"2024-01-14T03:18:00.26855","exception":false,"start_time":"2024-01-14T03:18:00.257628","status":"completed"},"tags":[]}},{"cell_type":"code","source":"class CFG:\n verbose = 1 # Verbosity\n seed = 42 # Random seed\n preset = \"efficientnetv2_b2_imagenet\" # Name of pretrained classifier\n image_size = [400, 300] # Input image size\n epochs = 5 # Training epochs\n batch_size = 64 # Batch size\n lr_mode = \"cos\" # LR scheduler mode from one of \"cos\", \"step\", \"exp\"\n drop_remainder = True # Drop incomplete batches\n num_classes = 6 # Number of classes in the dataset\n fold = 0 # Which fold to set as validation data\n class_names = ['Seizure', 'LPD', 'GPD', 'LRDA','GRDA', 'Other']\n label2name = dict(enumerate(class_names))\n name2label = {v:k for k, v in label2name.items()}","metadata":{"papermill":{"duration":0.018795,"end_time":"2024-01-14T03:18:00.298534","exception":false,"start_time":"2024-01-14T03:18:00.279739","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:33:59.407559Z","iopub.execute_input":"2024-09-12T17:33:59.407936Z","iopub.status.idle":"2024-09-12T17:33:59.414202Z","shell.execute_reply.started":"2024-09-12T17:33:59.4079Z","shell.execute_reply":"2024-09-12T17:33:59.413215Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# โ™ป๏ธ | Reproducibility \nSets value for random seed to produce similar result in each run.","metadata":{"papermill":{"duration":0.010907,"end_time":"2024-01-14T03:18:00.32063","exception":false,"start_time":"2024-01-14T03:18:00.309723","status":"completed"},"tags":[]}},{"cell_type":"code","source":"keras.utils.set_random_seed(CFG.seed)","metadata":{"papermill":{"duration":0.018371,"end_time":"2024-01-14T03:18:00.350074","exception":false,"start_time":"2024-01-14T03:18:00.331703","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:34:00.397431Z","iopub.execute_input":"2024-09-12T17:34:00.397857Z","iopub.status.idle":"2024-09-12T17:34:00.402807Z","shell.execute_reply.started":"2024-09-12T17:34:00.397819Z","shell.execute_reply":"2024-09-12T17:34:00.401626Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿ“ | Dataset Path ","metadata":{"papermill":{"duration":0.010888,"end_time":"2024-01-14T03:18:00.372053","exception":false,"start_time":"2024-01-14T03:18:00.361165","status":"completed"},"tags":[]}},{"cell_type":"code","source":"BASE_PATH = \"/kaggle/input/hms-harmful-brain-activity-classification\"\n\nSPEC_DIR = \"/tmp/dataset/hms-hbac\"\nos.makedirs(SPEC_DIR+'/train_spectrograms', exist_ok=True)\nos.makedirs(SPEC_DIR+'/test_spectrograms', exist_ok=True)","metadata":{"papermill":{"duration":0.017704,"end_time":"2024-01-14T03:18:00.400852","exception":false,"start_time":"2024-01-14T03:18:00.383148","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:34:06.612317Z","iopub.execute_input":"2024-09-12T17:34:06.612781Z","iopub.status.idle":"2024-09-12T17:34:06.619861Z","shell.execute_reply.started":"2024-09-12T17:34:06.612741Z","shell.execute_reply":"2024-09-12T17:34:06.619036Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿ“– | Meta Data ","metadata":{"papermill":{"duration":0.011434,"end_time":"2024-01-14T03:18:00.472401","exception":false,"start_time":"2024-01-14T03:18:00.460967","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# Train + Valid\ndf = pd.read_csv(f'{BASE_PATH}/train.csv')\ndf['eeg_path'] = f'{BASE_PATH}/train_eegs/'+df['eeg_id'].astype(str)+'.parquet'\ndf['spec_path'] = f'{BASE_PATH}/train_spectrograms/'+df['spectrogram_id'].astype(str)+'.parquet'\ndf['spec2_path'] = f'{SPEC_DIR}/train_spectrograms/'+df['spectrogram_id'].astype(str)+'.npy'\ndf['class_name'] = df.expert_consensus.copy()\ndf['class_label'] = df.expert_consensus.map(CFG.name2label)\ndisplay(df.head(2))\n\n# Test\ntest_df = pd.read_csv(f'{BASE_PATH}/test.csv')\ntest_df['eeg_path'] = f'{BASE_PATH}/test_eegs/'+test_df['eeg_id'].astype(str)+'.parquet'\ntest_df['spec_path'] = f'{BASE_PATH}/test_spectrograms/'+test_df['spectrogram_id'].astype(str)+'.parquet'\ntest_df['spec2_path'] = f'{SPEC_DIR}/test_spectrograms/'+test_df['spectrogram_id'].astype(str)+'.npy'\ndisplay(test_df.head(2))","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:34:08.668068Z","iopub.execute_input":"2024-09-12T17:34:08.668456Z","iopub.status.idle":"2024-09-12T17:34:09.303745Z","shell.execute_reply.started":"2024-09-12T17:34:08.668409Z","shell.execute_reply":"2024-09-12T17:34:09.302773Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Convert `.parquet` to `.npy`\n\nTo facilitate easier data loading, we will convert the EEG spectrograms from `parquet` to `npy` format. This process involves saving the spectrogram data, and since the content of the files remains the same, no significant changes are made. \n\n> It's worth noting that the `time` column is excluded, as it is not part of the spectrogram.","metadata":{}},{"cell_type":"code","source":"# Define a function to process a single eeg_id\ndef process_spec(spec_id, split=\"train\"):\n spec_path = f\"{BASE_PATH}/{split}_spectrograms/{spec_id}.parquet\"\n spec = pd.read_parquet(spec_path)\n spec = spec.fillna(0).values[:, 1:].T # fill NaN values with 0, transpose for (Time, Freq) -> (Freq, Time)\n spec = spec.astype(\"float32\")\n np.save(f\"{SPEC_DIR}/{split}_spectrograms/{spec_id}.npy\", spec)\n\n# Get unique spec_ids of train and valid data\nspec_ids = df[\"spectrogram_id\"].unique()\n\n# Parallelize the processing using joblib for training data\n_ = joblib.Parallel(n_jobs=-1, backend=\"loky\")(\n joblib.delayed(process_spec)(spec_id, \"train\")\n for spec_id in tqdm(spec_ids, total=len(spec_ids))\n)\n\n# Get unique spec_ids of test data\ntest_spec_ids = test_df[\"spectrogram_id\"].unique()\n\n# Parallelize the processing using joblib for test data\n_ = joblib.Parallel(n_jobs=-1, backend=\"loky\")(\n joblib.delayed(process_spec)(spec_id, \"test\")\n for spec_id in tqdm(test_spec_ids, total=len(test_spec_ids))\n)","metadata":{"papermill":{"duration":0.86264,"end_time":"2024-01-14T03:18:01.346487","exception":false,"start_time":"2024-01-14T03:18:00.483847","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:34:09.492754Z","iopub.execute_input":"2024-09-12T17:34:09.49363Z","iopub.status.idle":"2024-09-12T17:37:08.80459Z","shell.execute_reply.started":"2024-09-12T17:34:09.493591Z","shell.execute_reply":"2024-09-12T17:37:08.803683Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿš | DataLoader\n\nThis DataLoader first reads `npy` spectrogram files and extracts labeled subsamples using specified `offset` values. Then, it converts the spectrogram data into `log spectrogram` and applies the popular signal augmentation `MixUp`.\n\n> Note that, we are converting the mono channel signal to a 3-channel signal for using \"ImageNet\" weights of pretrained model.","metadata":{"papermill":{"duration":0.011843,"end_time":"2024-01-14T03:18:01.457956","exception":false,"start_time":"2024-01-14T03:18:01.446113","status":"completed"},"tags":[]}},{"cell_type":"code","source":"with tf.device('/GPU:0'):\n def build_augmenter(dim=CFG.image_size):\n augmenters = [\n keras_cv.layers.MixUp(alpha=2.0),\n keras_cv.layers.RandomCutout(height_factor=(1.0, 1.0),\n width_factor=(0.06, 0.1)), # freq-masking\n keras_cv.layers.RandomCutout(height_factor=(0.06, 0.1),\n width_factor=(1.0, 1.0)), # time-masking\n ]\n\n def augment(img, label):\n data = {\"images\":img, \"labels\":label}\n for augmenter in augmenters:\n if tf.random.uniform([]) < 0.5:\n data = augmenter(data, training=True)\n return data[\"images\"], data[\"labels\"]\n\n return augment\n\n\n def build_decoder(with_labels=True, target_size=CFG.image_size, dtype=32):\n def decode_signal(path, offset=None):\n # Read .npy files and process the signal\n file_bytes = tf.io.read_file(path)\n sig = tf.io.decode_raw(file_bytes, tf.float32)\n sig = sig[1024//dtype:] # Remove header tag\n sig = tf.reshape(sig, [400, -1])\n\n # Extract labeled subsample from full spectrogram using \"offset\"\n if offset is not None: \n offset = offset // 2 # Only odd values are given\n sig = sig[:, offset:offset+300]\n\n # Pad spectrogram to ensure the same input shape of [400, 300]\n pad_size = tf.math.maximum(0, 300 - tf.shape(sig)[1])\n sig = tf.pad(sig, [[0, 0], [0, pad_size]])\n sig = tf.reshape(sig, [400, 300])\n\n # Log spectrogram \n sig = tf.clip_by_value(sig, tf.math.exp(-4.0), tf.math.exp(8.0)) # avoid 0 in log\n sig = tf.math.log(sig)\n\n # Normalize spectrogram\n sig -= tf.math.reduce_mean(sig)\n sig /= tf.math.reduce_std(sig) + 1e-6\n\n # Mono channel to 3 channels to use \"ImageNet\" weights\n sig = tf.tile(sig[..., None], [1, 1, 3])\n return sig\n\n def decode_label(label):\n label = tf.one_hot(label, CFG.num_classes)\n label = tf.cast(label, tf.float32)\n label = tf.reshape(label, [CFG.num_classes])\n return label\n\n def decode_with_labels(path, offset=None, label=None):\n sig = decode_signal(path, offset)\n label = decode_label(label)\n return (sig, label)\n\n return decode_with_labels if with_labels else decode_signal\n\n\n def build_dataset(paths, offsets=None, labels=None, batch_size=32, cache=True,\n decode_fn=None, augment_fn=None,\n augment=False, repeat=True, shuffle=1024, \n cache_dir=\"\", drop_remainder=False):\n if cache_dir != \"\" and cache is True:\n os.makedirs(cache_dir, exist_ok=True)\n\n if decode_fn is None:\n decode_fn = build_decoder(labels is not None)\n\n if augment_fn is None:\n augment_fn = build_augmenter()\n\n AUTO = tf.data.experimental.AUTOTUNE\n slices = (paths, offsets) if labels is None else (paths, offsets, labels)\n\n ds = tf.data.Dataset.from_tensor_slices(slices)\n ds = ds.map(decode_fn, num_parallel_calls=AUTO)\n ds = ds.cache(cache_dir) if cache else ds\n ds = ds.repeat() if repeat else ds\n if shuffle: \n ds = ds.shuffle(shuffle, seed=CFG.seed)\n opt = tf.data.Options()\n opt.experimental_deterministic = False\n ds = ds.with_options(opt)\n ds = ds.batch(batch_size, drop_remainder=drop_remainder)\n ds = ds.map(augment_fn, num_parallel_calls=AUTO) if augment else ds\n ds = ds.prefetch(AUTO)\n return ds","metadata":{"papermill":{"duration":0.039133,"end_time":"2024-01-14T03:18:01.509017","exception":false,"start_time":"2024-01-14T03:18:01.469884","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:08.806547Z","iopub.execute_input":"2024-09-12T17:37:08.806871Z","iopub.status.idle":"2024-09-12T17:37:08.827701Z","shell.execute_reply.started":"2024-09-12T17:37:08.806838Z","shell.execute_reply":"2024-09-12T17:37:08.826827Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿ”ช | Data Split\n\nIn the following code snippet, the data is divided into `5` folds. Note that, the `groups` argument is used to prevent any overlap of patients between the training and validation sets, thus avoiding potential **data leakage** issues. Additionally, each split is stratified based on the `class_label`, ensuring a uniform distribution of class labels in each fold.","metadata":{"papermill":{"duration":0.012174,"end_time":"2024-01-14T03:18:01.538524","exception":false,"start_time":"2024-01-14T03:18:01.52635","status":"completed"},"tags":[]}},{"cell_type":"code","source":"from sklearn.model_selection import StratifiedGroupKFold\n\nsgkf = StratifiedGroupKFold(n_splits=5, shuffle=True, random_state=CFG.seed)\n\ndf[\"fold\"] = -1\ndf.reset_index(drop=True, inplace=True)\nfor fold, (train_idx, valid_idx) in enumerate(\n sgkf.split(df, y=df[\"class_label\"], groups=df[\"patient_id\"])\n):\n df.loc[valid_idx, \"fold\"] = fold\ndf.groupby([\"fold\", \"class_name\"])[[\"eeg_id\"]].count().T","metadata":{"papermill":{"duration":0.037496,"end_time":"2024-01-14T03:18:01.587924","exception":false,"start_time":"2024-01-14T03:18:01.550428","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:08.828913Z","iopub.execute_input":"2024-09-12T17:37:08.829196Z","iopub.status.idle":"2024-09-12T17:37:10.771359Z","shell.execute_reply.started":"2024-09-12T17:37:08.829165Z","shell.execute_reply":"2024-09-12T17:37:10.770492Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Build Train & Valid Dataset\n\nOnly first sample for each `spectrogram_id` is used in order to keep the dataset size managable. Feel free to train on full data.","metadata":{"papermill":{"duration":0.011875,"end_time":"2024-01-14T03:18:01.611955","exception":false,"start_time":"2024-01-14T03:18:01.60008","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"# Sample from full data\nsample_df = df.groupby(\"spectrogram_id\").head(1).reset_index(drop=True)\ntrain_df = sample_df[sample_df.fold != CFG.fold]\nvalid_df = sample_df[sample_df.fold == CFG.fold]\nprint(f\"# Num Train: {len(train_df)} | Num Valid: {len(valid_df)}\")\n\n# Train\ntrain_paths = train_df.spec2_path.values\ntrain_offsets = train_df.spectrogram_label_offset_seconds.values.astype(int)\ntrain_labels = train_df.class_label.values\ntrain_ds = build_dataset(train_paths, train_offsets, train_labels, batch_size=CFG.batch_size,\n repeat=True, shuffle=True, augment=True, cache=True)\n\n# Valid\nvalid_paths = valid_df.spec2_path.values\nvalid_offsets = valid_df.spectrogram_label_offset_seconds.values.astype(int)\nvalid_labels = valid_df.class_label.values\nvalid_ds = build_dataset(valid_paths, valid_offsets, valid_labels, batch_size=CFG.batch_size,\n repeat=False, shuffle=False, augment=False, cache=True)","metadata":{"execution":{"iopub.status.busy":"2024-09-12T15:57:02.463946Z","iopub.execute_input":"2024-09-12T15:57:02.464974Z","iopub.status.idle":"2024-09-12T15:57:05.412746Z","shell.execute_reply.started":"2024-09-12T15:57:02.464922Z","shell.execute_reply":"2024-09-12T15:57:05.411933Z"}}},{"cell_type":"code","source":"# Sample from full data\nsample_df = df.groupby(\"spectrogram_id\").head(1).reset_index(drop=True)\n\n# Split the data into train, validation, and test\ntrain_df = sample_df[sample_df.fold != CFG.fold]\nvalid_df = sample_df[sample_df.fold == CFG.fold]\ntest_df = sample_df[sample_df.fold == (CFG.fold + 1) % sample_df.fold.nunique()] # Use a different fold for testing\n\nprint(f\"# Num Train: {len(train_df)} | Num Valid: {len(valid_df)} | Num Test: {len(test_df)}\")\n\n# Train\ntrain_paths = train_df.spec2_path.values\ntrain_offsets = train_df.spectrogram_label_offset_seconds.values.astype(int)\ntrain_labels = train_df.class_label.values\ntrain_ds = build_dataset(train_paths, train_offsets, train_labels, batch_size=CFG.batch_size,\n repeat=True, shuffle=True, augment=True, cache=True)\n\n# Valid\nvalid_paths = valid_df.spec2_path.values\nvalid_offsets = valid_df.spectrogram_label_offset_seconds.values.astype(int)\nvalid_labels = valid_df.class_label.values\nvalid_ds = build_dataset(valid_paths, valid_offsets, valid_labels, batch_size=CFG.batch_size,\n repeat=False, shuffle=False, augment=False, cache=True)\n\n# Test\ntest_paths = test_df.spec2_path.values\ntest_offsets = test_df.spectrogram_label_offset_seconds.values.astype(int)\ntest_labels = test_df.class_label.values\ntest_ds = build_dataset(test_paths, test_offsets, test_labels, batch_size=CFG.batch_size,\n repeat=False, shuffle=False, augment=False, cache=True)\n","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:37:10.773788Z","iopub.execute_input":"2024-09-12T17:37:10.774283Z","iopub.status.idle":"2024-09-12T17:37:13.445421Z","shell.execute_reply.started":"2024-09-12T17:37:10.774248Z","shell.execute_reply":"2024-09-12T17:37:13.444622Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Dataset Check\n\nLet's visualize some samples from the dataset.","metadata":{}},{"cell_type":"code","source":"imgs, tars = next(iter(train_ds))\n\nnum_imgs = 8\nplt.figure(figsize=(4*4, num_imgs//4*5))\nfor i in range(num_imgs):\n plt.subplot(num_imgs//4, 4, i + 1)\n img = imgs[i].numpy()[...,0] # Adjust as per your image data format\n img -= img.min()\n img /= img.max() + 1e-4\n tar = CFG.label2name[np.argmax(tars[i].numpy())]\n plt.imshow(img)\n plt.title(f\"Target: {tar}\")\n plt.axis('off')\n \nplt.tight_layout()\nplt.show()","metadata":{"_kg_hide-input":true,"execution":{"iopub.status.busy":"2024-09-12T17:37:13.44658Z","iopub.execute_input":"2024-09-12T17:37:13.446884Z","iopub.status.idle":"2024-09-12T17:37:16.327321Z","shell.execute_reply.started":"2024-09-12T17:37:13.446852Z","shell.execute_reply":"2024-09-12T17:37:16.326019Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿ” | Loss & Metric\n\nThe evaluation metric in this competition is **KL Divergence**, defined as,\n\n$$\nD_{\\text{KL}}(P \\parallel Q) = \\sum_{i} P(i) \\log\\left(\\frac{P(i)}{Q(i)}\\right)\n$$\n\nWhere:\n- $P$ is the true distribution.\n- $Q$ is the predicted distribution.\n\nInterestingly, as KL Divergence is differentiable, we can directly use it as our loss function. Thus, we don't need to use a third-party metric like **Accuracy** to evaluate our model. Therefore, `valid_loss` can stand alone as an indicator for our evaluation. In keras, we already have impelementation for KL Divergence loss so we only need to import it.","metadata":{}},{"cell_type":"code","source":"LOSS = keras.losses.KLDivergence()","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:37:16.328953Z","iopub.execute_input":"2024-09-12T17:37:16.329257Z","iopub.status.idle":"2024-09-12T17:37:16.333342Z","shell.execute_reply.started":"2024-09-12T17:37:16.329225Z","shell.execute_reply":"2024-09-12T17:37:16.33258Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿค– | Modeling\n\nThis notebook uses the `EfficientNetV2 B2` from KerasCV's collection of pretrained models. To explore other models, simply modify the `preset` in the `CFG` (config). Check the [KerasCV website](https://keras.io/api/keras_cv/models/tasks/image_classifier/) for a list of available pretrained models.","metadata":{"papermill":{"duration":0.016849,"end_time":"2024-01-14T03:18:38.613991","exception":false,"start_time":"2024-01-14T03:18:38.597142","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# Build Classifier\nwith tf.device('/GPU:0'):\n model = keras_cv.models.ImageClassifier.from_preset(\n CFG.preset, num_classes=CFG.num_classes\n )\n\n # Compile the model \n model.compile(optimizer=keras.optimizers.Adam(learning_rate=1e-4),\n loss=LOSS,metrics=['accuracy'])\n\n # Model Sumamry\n model.summary()","metadata":{"papermill":{"duration":10.446166,"end_time":"2024-01-14T03:18:49.186176","exception":false,"start_time":"2024-01-14T03:18:38.74001","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:16.334633Z","iopub.execute_input":"2024-09-12T17:37:16.33504Z","iopub.status.idle":"2024-09-12T17:37:33.955489Z","shell.execute_reply.started":"2024-09-12T17:37:16.334992Z","shell.execute_reply":"2024-09-12T17:37:33.954584Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# โš“ | LR Schedule\n\nA well-structured learning rate schedule is essential for efficient model training, ensuring optimal convergence and avoiding issues such as overshooting or stagnation.","metadata":{"papermill":{"duration":0.016209,"end_time":"2024-01-14T03:18:49.21924","exception":false,"start_time":"2024-01-14T03:18:49.203031","status":"completed"},"tags":[]}},{"cell_type":"code","source":"import math\n\ndef get_lr_callback(batch_size=8, mode='cos', epochs=10, plot=False):\n lr_start, lr_max, lr_min = 5e-5, 6e-6 * batch_size, 1e-5\n lr_ramp_ep, lr_sus_ep, lr_decay = 3, 0, 0.75\n\n def lrfn(epoch): # Learning rate update function\n if epoch < lr_ramp_ep: lr = (lr_max - lr_start) / lr_ramp_ep * epoch + lr_start\n elif epoch < lr_ramp_ep + lr_sus_ep: lr = lr_max\n elif mode == 'exp': lr = (lr_max - lr_min) * lr_decay**(epoch - lr_ramp_ep - lr_sus_ep) + lr_min\n elif mode == 'step': lr = lr_max * lr_decay**((epoch - lr_ramp_ep - lr_sus_ep) // 2)\n elif mode == 'cos':\n decay_total_epochs, decay_epoch_index = epochs - lr_ramp_ep - lr_sus_ep + 3, epoch - lr_ramp_ep - lr_sus_ep\n phase = math.pi * decay_epoch_index / decay_total_epochs\n lr = (lr_max - lr_min) * 0.5 * (1 + math.cos(phase)) + lr_min\n return lr\n\n if plot: # Plot lr curve if plot is True\n plt.figure(figsize=(10, 5))\n plt.plot(np.arange(epochs), [lrfn(epoch) for epoch in np.arange(epochs)], marker='o')\n plt.xlabel('epoch'); plt.ylabel('lr')\n plt.title('LR Scheduler')\n plt.show()\n\n return keras.callbacks.LearningRateScheduler(lrfn, verbose=False) # Create lr callback","metadata":{"papermill":{"duration":0.028945,"end_time":"2024-01-14T03:18:49.264535","exception":false,"start_time":"2024-01-14T03:18:49.23559","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:33.956777Z","iopub.execute_input":"2024-09-12T17:37:33.957159Z","iopub.status.idle":"2024-09-12T17:37:33.968279Z","shell.execute_reply.started":"2024-09-12T17:37:33.957115Z","shell.execute_reply":"2024-09-12T17:37:33.967402Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"lr_cb = get_lr_callback(CFG.batch_size, mode=CFG.lr_mode, plot=True)","metadata":{"papermill":{"duration":0.297147,"end_time":"2024-01-14T03:18:49.578089","exception":false,"start_time":"2024-01-14T03:18:49.280942","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:33.969591Z","iopub.execute_input":"2024-09-12T17:37:33.970176Z","iopub.status.idle":"2024-09-12T17:37:34.183758Z","shell.execute_reply.started":"2024-09-12T17:37:33.970133Z","shell.execute_reply":"2024-09-12T17:37:34.182777Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿ’พ | Model Checkpointing","metadata":{"papermill":{"duration":0.017199,"end_time":"2024-01-14T03:18:49.613648","exception":false,"start_time":"2024-01-14T03:18:49.596449","status":"completed"},"tags":[]}},{"cell_type":"code","source":"ckpt_cb = keras.callbacks.ModelCheckpoint(\"best_model.keras\",\n monitor='val_loss',\n save_best_only=True,\n save_weights_only=False,\n mode='min')","metadata":{"papermill":{"duration":0.024529,"end_time":"2024-01-14T03:18:49.655708","exception":false,"start_time":"2024-01-14T03:18:49.631179","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:34.186607Z","iopub.execute_input":"2024-09-12T17:37:34.186885Z","iopub.status.idle":"2024-09-12T17:37:34.191506Z","shell.execute_reply.started":"2024-09-12T17:37:34.186855Z","shell.execute_reply":"2024-09-12T17:37:34.190513Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿš‚ | Training","metadata":{"papermill":{"duration":0.01671,"end_time":"2024-01-14T03:18:49.689354","exception":false,"start_time":"2024-01-14T03:18:49.672644","status":"completed"},"tags":[]}},{"cell_type":"code","source":"with tf.device('/GPU:0'):\n history = model.fit(\n train_ds, \n epochs=CFG.epochs,\n callbacks=[lr_cb, ckpt_cb], \n steps_per_epoch=len(train_df)//CFG.batch_size,\n validation_data=valid_ds, \n verbose=CFG.verbose\n )","metadata":{"papermill":{"duration":3374.692199,"end_time":"2024-01-14T04:15:04.398389","exception":false,"start_time":"2024-01-14T03:18:49.70619","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:37:34.192761Z","iopub.execute_input":"2024-09-12T17:37:34.193123Z","iopub.status.idle":"2024-09-12T17:45:10.902547Z","shell.execute_reply.started":"2024-09-12T17:37:34.193082Z","shell.execute_reply":"2024-09-12T17:45:10.901468Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"# ๐Ÿงช | Prediction","metadata":{"papermill":{"duration":0.693309,"end_time":"2024-01-14T04:15:05.731839","exception":false,"start_time":"2024-01-14T04:15:05.03853","status":"completed"},"tags":[]}},{"cell_type":"markdown","source":"## Load Best Model","metadata":{"papermill":{"duration":0.632183,"end_time":"2024-01-14T04:15:06.991143","exception":false,"start_time":"2024-01-14T04:15:06.35896","status":"completed"},"tags":[]}},{"cell_type":"code","source":"model.load_weights('/kaggle/working/best_model.keras')\nimport joblib\n\n# Assuming you have a trained model named `model`\nmodel_save_path = 'my_model3.joblib'\n\n# Save the model using joblib\njoblib.dump(model, model_save_path)\n\nprint(f\"Model saved to {model_save_path}\")","metadata":{"papermill":{"duration":20.428261,"end_time":"2024-01-14T04:15:28.044401","exception":false,"start_time":"2024-01-14T04:15:07.61614","status":"completed"},"tags":[],"execution":{"iopub.status.busy":"2024-09-12T17:45:10.904242Z","iopub.execute_input":"2024-09-12T17:45:10.904551Z","iopub.status.idle":"2024-09-12T17:45:19.697024Z","shell.execute_reply.started":"2024-09-12T17:45:10.904519Z","shell.execute_reply":"2024-09-12T17:45:19.696096Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Build Test Dataset","metadata":{"papermill":{"duration":0.703901,"end_time":"2024-01-14T04:20:09.745279","exception":false,"start_time":"2024-01-14T04:20:09.041378","status":"completed"},"tags":[]}},{"cell_type":"code","source":"# Evaluate the model on the test dataset\ntest_loss, test_accuracy = model.evaluate(\n test_ds, # Test dataset\n verbose=CFG.verbose # Verbosity of the output\n)\n\n# Print the results\nprint(f\"Test Loss: {test_loss:.4f}\")\nprint(f\"Test Accuracy: {test_accuracy:.4f}\")","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:45:19.698173Z","iopub.execute_input":"2024-09-12T17:45:19.698513Z","iopub.status.idle":"2024-09-12T17:45:41.82491Z","shell.execute_reply.started":"2024-09-12T17:45:19.69848Z","shell.execute_reply":"2024-09-12T17:45:41.823987Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"import numpy as np\nimport pandas as pd\nfrom sklearn.metrics import confusion_matrix, classification_report\nimport seaborn as sns\nimport matplotlib.pyplot as plt\n\n# Obtain predictions\npredictions = model.predict(test_ds)\npredicted_labels = np.argmax(predictions, axis=-1) # Get the class with the highest probability\n\n# Extract true labels from the test dataset\ntrue_labels = []\nfor _, labels in test_ds:\n true_labels.extend(np.argmax(labels, axis=-1))\ntrue_labels = np.array(true_labels)\n\n# Compute confusion matrix\ncm = confusion_matrix(true_labels, predicted_labels, labels=np.arange(CFG.num_classes))\n\n# Plot confusion matrix\nplt.figure(figsize=(10, 7))\nsns.heatmap(cm, annot=True, fmt='d', cmap='Blues', \n xticklabels=CFG.class_names, yticklabels=CFG.class_names)\nplt.xlabel('Predicted Label')\nplt.ylabel('True Label')\nplt.title('Confusion Matrix')\nplt.show()\n\n# Generate classification report\nreport = classification_report(true_labels, predicted_labels, target_names=CFG.class_names)\nprint(report)","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:45:41.826175Z","iopub.execute_input":"2024-09-12T17:45:41.826528Z","iopub.status.idle":"2024-09-12T17:46:03.09935Z","shell.execute_reply.started":"2024-09-12T17:45:41.826479Z","shell.execute_reply":"2024-09-12T17:46:03.098409Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"predictions = model.predict(test_ds)\npredicted_labels = np.argmax(predictions, axis=-1)","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:46:03.100523Z","iopub.execute_input":"2024-09-12T17:46:03.10112Z","iopub.status.idle":"2024-09-12T17:46:07.010915Z","shell.execute_reply.started":"2024-09-12T17:46:03.101084Z","shell.execute_reply":"2024-09-12T17:46:07.009937Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"report = classification_report(true_labels, predicted_labels, target_names=CFG.class_names)\nprint(report)","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:46:07.011923Z","iopub.execute_input":"2024-09-12T17:46:07.012215Z","iopub.status.idle":"2024-09-12T17:46:07.026241Z","shell.execute_reply.started":"2024-09-12T17:46:07.012185Z","shell.execute_reply":"2024-09-12T17:46:07.025486Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"markdown","source":"## Inference","metadata":{}},{"cell_type":"code","source":"plt.plot(history.history['loss'],label='Train_loss')\nplt.plot(history.history['val_loss'],label='Val_loss')\nplt.legend()\nplt.xlabel('No. of Epochs')\nplt.ylabel('Loss')\nplt.title('Loss vs Epochs')\nplt.show()\n\n\nplt.plot(history.history['accuracy'],label = 'Train_acc')\nplt.plot(history.history['val_accuracy'],label = 'Val_acc')\nplt.legend()\nplt.xlabel('No. of Epochs')\nplt.ylabel(\"Accuracy\")\nplt.title(\"Accuracy vs Epochs\")\nplt.show()","metadata":{"execution":{"iopub.status.busy":"2024-09-12T17:46:07.027313Z","iopub.execute_input":"2024-09-12T17:46:07.027603Z","iopub.status.idle":"2024-09-12T17:46:07.615696Z","shell.execute_reply.started":"2024-09-12T17:46:07.027572Z","shell.execute_reply":"2024-09-12T17:46:07.614719Z"},"trusted":true},"execution_count":null,"outputs":[]},{"cell_type":"code","source":"","metadata":{},"execution_count":null,"outputs":[]}]}