Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Add a ShardDatasetBuilder that creates shards directly. #5605

Merged
merged 1 commit into from
Sep 24, 2024
Merged

Add a ShardDatasetBuilder that creates shards directly. #5605

merged 1 commit into from
Sep 24, 2024

Commits on Sep 24, 2024

  1. Add a ShardDatasetBuilder that creates shards directly. 

    In certain cases, users have data available in different shards and they want to keep the same number of shards and in each shard the same order of examples (or they don't care about the ordering). In that case, our current dataset builder classes are much slower than necessary.

The `ShardBasedBuilder` allows users to create dataset builders that process source data shard by shard. It can be run with or without Beam. In case of Beam, the resulting Beam pipeline is significantly simpler and therefore faster.

PiperOrigin-RevId: 678137550
    @tomvdw
    tomvdw authored and The TensorFlow Datasets Authors committed Sep 24, 2024
    Configuration menu
    Copy the full SHA
    8223a15 View commit details
    Browse the repository at this point in the history