{"cells":[{"metadata":{"_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","trusted":true},"cell_type":"code","source":"import os\nimport shutil\nimport pandas as pd\nimport tensorflow as tf\nfrom kaggle_datasets import KaggleDatasets\n\nGCS_PATH = KaggleDatasets().get_gcs_path('ranzcr-clip-catheter-line-classification')\nTRAINING_FILENAMES = tf.io.gfile.glob(GCS_PATH + '/train_tfrecords/*.tfrec')\n\ndef read_labeled_tfrecord(example):\n    example = tf.io.parse_single_example(example, {\"StudyInstanceUID\": tf.io.FixedLenFeature([], tf.string)})\n    return example['StudyInstanceUID']","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# read folds and create dirs\nfolds = pd.read_csv('../input/ranzcr-folds/folds.csv')\nfolds = dict(zip(folds['StudyInstanceUID'],folds['fold']))\nfor i in range(5):\n    os.mkdir(f'fold{i}')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"for i in range(len(TRAINING_FILENAMES)):\n    dataset = tf.data.TFRecordDataset(TRAINING_FILENAMES[i])\n    for batch in dataset.batch(1):\n        batch_ds = tf.data.Dataset.from_tensor_slices([*batch])\n        batch_rec = batch_ds.map(read_labeled_tfrecord)\n        for x in batch_rec:\n            fname = str(tf.keras.backend.eval(x).decode(\"utf-8\"))\n        fold = folds[fname]\n        filename = f\"fold{fold}/{fname}.tfrecord\"\n        writer = tf.data.experimental.TFRecordWriter(filename)\n        writer.write(batch_ds)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# make single tfrec file for each fold\nfor i in range(5):\n    list_of_tfrecord_files = os.listdir(f\"fold{i}/\")\n    dataset = tf.data.TFRecordDataset([f\"fold{i}/\"+x for x in list_of_tfrecord_files])\n    filename = f'fold{i}-{len(list_of_tfrecord_files)}.tfrec'\n    writer = tf.data.experimental.TFRecordWriter(filename)\n    writer.write(dataset)\n    shutil.rmtree(f'fold{i}')\n                                        \nprint(\"Done!\")                                        ","execution_count":null,"outputs":[]}],"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":4,"nbformat_minor":4}