v0.2.40
What's Changed
- fix:
clean parquet dir cachefixture by @deependujha in #474 - Fix: Allow using
Machinetypes inmapby @ethanwharris in #473 - 🛠️ Fix: Ensure
chunk_bytesinindex.jsonmatches actual chunk file size by @bhimrazy in #478 - fix: _get_folder_size fn by @deependujha in #471
- Added boolean serialiser called by litdata.optimise() by @DominiquePaul in #481
- Doc: improve dev doc & add ToDos by @deependujha in #479
- upd: Add hf file download progress and update local file path by @bhimrazy in #484
- fix: segmentation fault error in streaming tokens by @bhimrazy in #485
- Warn user if
max_cache_sizeis less than 25GB in StreamingDataset by @bhimrazy in #489 - fix: Properly assign the chunks to the right worker by @tchaton in #449
- Bump version to 0.2.40 by @bhimrazy in #491
New Contributors
- @ethanwharris made their first contribution in #473
- @DominiquePaul made their first contribution in #481
Full Changelog: v0.2.39...v0.2.40