Skip to content

IBM Dataloader - #19

Open
ProgerDav wants to merge 7 commits into
mainfrom
daviswer_main
Open

IBM Dataloader#19
ProgerDav wants to merge 7 commits into
mainfrom
daviswer_main

Conversation

@ProgerDav

Copy link
Copy Markdown

No description provided.

daviswer and others added 4 commits May 31, 2024 03:24
Co-authored-by: Linsong Chu <lchu@us.ibm.com>
Re-sync fork to main, and add the latest features from fms-fsdp:

- Remove dependence on the separate metadata "count file" in the dataset directory
- Support n_workers > 1. This is accomplished by shunting all path/rank-dependent setup out of initialization and into a new setup() method, which runs after init but before any other op
- Support HF-style parquet raw text datasets, with tokenization on the fly (for reasonably sized documents/shardfiles)
- Support non-flat data directories: all legal files under the specified location will be included, regardless of depth or location. Enables simple weight-free dataset mixing via a single `StreamingDocDataset` on the parent directory
- `SamplingDataset` and `ScalableShardDataset`are now implemented as proper `_WrapperDatasets`, reflecting the intended modular usage
- Fix Weird_Separated_Camel_Case naming convention in favor of ProperClassNaming
- Allow `PreloadBufferDataset` to shrink back down to the desired size after rescaling to a smaller number of workers
@philippguevorguian
philippguevorguian marked this pull request as draft September 10, 2024 12:02
@ProgerDav
ProgerDav marked this pull request as ready for review September 10, 2024 13:39
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants