Skip to content

Conversation

Abdul-Omira
Copy link

Add a new 'Streaming best practices' page with practical patterns and pitfalls for large-scale/production use of IterableDataset. Includes examples for batched map with remove_columns, deterministic shuffling with set_epoch, multi-worker sharding, checkpoint/resume, and persistence to Parquet/Hub. Linked from How-to > General usage, next to Stream.

…alls, and examples; link from Stream section
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.

1 participant