huggingface / huggingface/datasets

adding progress bar / ETA for `load_dataset`

Open
#2,649 2 comments 7 reactions 0 assignees View on GitHub
enhancement
Dominant language
Python
Stars
22k
Forks
3.4k
Avg merge
5d 7h
Merged PRs (30d)
17

Description

Please consider:
```
Downloading and preparing dataset oscar/unshuffled_deduplicated_en (download: 462.40 GiB, generated: 1.18 TiB, post-processed: Unknown size, total: 1.63 TiB) to cache/oscar/unshuffled_deduplicated_en/1.0.0/84838bd49d2295f62008383b05620571535451d84545037bb94d6f3501651df2...
HF google storage unreachable. Downloading and preparing it from source
```
and no indication whatsoever of whether things work well or when it'll be done. It's important to have an estimated completion time for when doing slurm jobs since some instances have a cap on run-time.

I think for this particular job it sat for 30min in total silence and then after 30min it started generating:
```
897850 examples [07:24, 10286.71 examples/s]
```
which is already great!

Request:
1. ETA - knowing how many hours to allocate for a slurm job
2. progress bar - helps to know things are working and aren't stuck and where we are at.

Thank you!

@lhoestq

Contributor guide

Open the contributing guide

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.