imdb source error

[Muhtasham]

Describe the bug

imdb dataset not loading

Steps to reproduce the bug

from datasets import load_dataset
dataset = load_dataset("imdb")

Expected results

Actual results

06/23/2022 14:45:18 - INFO - datasets.builder - Dataset not on Hf google storage. Downloading and preparing it from source
06/23/2022 14:46:34 - INFO - datasets.utils.file_utils - HEAD request to http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz timed out, retrying... [1.0]
.....
ConnectionError: Couldn't reach http://ai.stanford.edu/~amaas/data/sentiment/aclImdb_v1.tar.gz (ConnectTimeout(MaxRetryError("HTTPConnectionPool(host='ai.stanford.edu', port=80): Max retries exceeded with url: /~amaas/data/sentiment/aclImdb_v1.tar.gz (Caused by ConnectTimeoutError(<urllib3.connection.HTTPConnection object at 0x7f2d750cf690>, 'Connection to ai.stanford.edu timed out. (connect timeout=100)'))")))

Environment info

• datasets version: 2.3.2
• Platform: Linux-5.4.188+-x86_64-with-Ubuntu-18.04-bionic
• Python version: 3.7.13
• PyArrow version: 6.0.1
• Pandas version: 1.3.5

[albertvillanova]

Thanks for reporting, @Muhtasham.

Indeed IMDB dataset is not accessible from yesterday, because the data is hosted on the data owners servers at Stanford (http://ai.stanford.edu/) and these are down due to a power outage originated by a fire: https://twitter.com/StanfordAILab/status/1539472302399623170?s=20&t=1HU1hrtaXprtn14U61P55w

As a temporary workaroud, you can load the IMDB dataset with this tweak:

ds = load_dataset("imdb", revision="tmp-fix-imdb")