Context This is a huge dataset and takes around 400 seconds to load into kernel. If you need quickly IMDB data in Keras kernel use the following dataset instead: [https://www.kaggle.com/pankrzysiu/keras-imdb-reviews][1] Content A set of 50,000 highly-polarized reviews from the Internet Movie Database. Usage Instructions # aclImdb_v1.zip file This file is to be used directly in your code. The .zip file will be automatically uncompressed by Kaggle. # imdb* files from os import listdir, makedirs from os.path import join, exists, expanduser cache_dir = expanduser(join('~', '.keras')) if not exists(cache_dir): makedirs(cache_dir) datasets_dir = join(cache_dir, 'datasets') if not exists(datasets_dir): makedirs(datasets_dir) # If you have multiple input files, change the below cp commands accordingly, typically: # !cp ../input/keras-imdb/imdb* ~/.keras/datasets/ !cp ../input/imdb* ~/.keras/datasets/ Acknowledgements The files are on the net in these locations: [https://s3.amazonaws.com/text-datasets/imdb.npz][2] [https://s3.amazonaws.com/text-datasets/imdb_word_index.json][3] They are used by keras imdb.py: [https://github.com/keras-team/keras/blob/master/keras/datasets/imdb.py][4] Inspiration "Python Deep Learning" Book example is using this: https://github.com/fchollet/deep-learning-with-python-notebooks/blob/master/6.1-using-word-embeddings.ipynb [1]: https://www.kaggle.com/pankrzysiu/keras-imdb-reviews [2]: https://s3.amazonaws.com/text-datasets/imdb.npz [3]: https://s3.amazonaws.com/text-datasets/imdb_word_index.json [4]: https://github.com/keras-team/keras/blob/master/keras/datasets/imdb.py