[Examples] Add automatic dataset splitting in language-modeling examples #9133
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
|
Ah, the commit from #9127 seems to have snuck its way in there. Should I remove it? |
|
If you can do it easily, that would be best! |
sgugger
approved these changes
Dec 15, 2020
I've tried for a bit but I think I just made things worse ! If that's OK I'll leave it there and I'll fix things at merge time. |
guyrosin
pushed a commit
to guyrosin/transformers
that referenced
this pull request
Jan 15, 2021
…les (huggingface#9133) * replaced jnp.split + removing textual model inputs + ensuring warmup_steps > 0 * Add automatic dataset splitting in language-modeling examples
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What does this PR do?
Currently, language-modeling examples support passing a HF-datasets dataset as training data. However, this dataset needs to have a
trainandvalidationsplit, which is not the case for many language-modeling datasets, which are just unstructured text. The updated scripts automatically partition thetrainsplit to create avalidationsplit if it doesn't exist already, and addsvalidation_split_percentageargument to control the split ratio, set to 5% by default.