| import tensorflow_datasets as tfds | |
| import os | |
| cfg = tfds.download.DownloadConfig(extract_dir=os.getcwd() + '/raw_datasets/') | |
| tfds.load('speech_commands', data_dir='./raw_datasets', download=True, download_and_prepare_kwargs={'download_dir': os.getcwd() + '/raw_datasets/', 'download_config': cfg}) | |