Datasets:

cis-lmu

/

GlotCC-V1

like 33

Now the state changed to:

The dataset viewer is taking too long to fetch the data.

Error code:   ClientConnectionError

again the same as before

Error details:

Error code:   ResponseNotFound

refreshing manually

Will this do the trick?

Seems like it started to show subsets... we just need to wait.

It works for some of the langauges, but now it has a new error for some others

Error details:

Error code:   LockedDatasetTimeoutError

cc @severo .

Ok it seems some are failed, is it possbile to run the failed part again?

https://datasets-server.huggingface.co/parquet?dataset=cis-lmu/GlotCC-V1

cc @albertvillanova @lhoestq @severo .

Error code: LockedDatasetTimeoutError

these errors are retried automatically after some time, nothing special to do.

If I reviewed well, everything is working now on this dataset!

@severo Thanks so much. Now I’m afraid that if I change the content of the README, it will build it again from the start and not become like this again.

Yes, while https://github.com/huggingface/dataset-viewer/pull/2788 is still in progress, any update will retrigger the viewer refresh. cc @lhoestq for visibility. Feel free to to it anyway if you need to, and to report here if it does not work, but sure: for big datasets, it's better to have few commits.

Is it possible to show the dataset viewer for all subsets instead of just for the first 5 GB of each subset?

cc @severo

Is it possible to show the dataset viewer for all subsets instead of just for the first 5 GB of each subset?

Not at the moment. Feel free to upvote or comment on this issue: https://github.com/huggingface/dataset-viewer/issues/2878