Bump grain from 0.2.16 to 0.2.17

[dependabot]

Bumps grain from 0.2.16 to 0.2.17.

Release notes

Sourced from grain's releases.

Grain 0.2.17

• New features:

  • Adds automated batching into shared memory when multiprocess prefetch is used to save one data copy.
  • Updates TfMixtureIndexSampler to accept dynamic 1D/2D Tensor/Variable proportions that could be updated during runtime.
  • Automatic Xprof subprocess trace collection into PyGrain workers. To enable, set flag grain_enable_multiprocess_worker_profiling=true and add "profile_subprocesses" = True in advanced profiler options.

• Breaking changes:

  • Force index shuffle rounds to not be bigger than 1024

• Bug fixes:

  • Preserve and reconstruct tracebacks for exceptions thrown in worker processes across multiprocessing boundaries
  • Fixes performance regression in traceback filtering when workers are frequently exhausting small datasets.
  • Fixes silent string truncation in shared memory batching.

Changelog

Sourced from grain's changelog.

Change log

The change log file hosting all releases with lists of new features and breaking changes. Best viewed here.

Unreleased

• New features:

  • Adds automated batching into shared memory when multiprocess prefetch is used to save one data copy.
  • Adds experimental ShapeDtypeStructProtocol and ShapeDtypeStruct to represent dataset element specs.
  • Updates TfMixtureIndexSampler to accept dynamic 1D/2D Tensor/Variable proportions that could be updated during runtime.
  • Adds profiling of multiprocess workers when using XProf profiler. To enable, set flag grain_enable_multiprocess_worker_profiling=true and add "profile_subprocesses" = True in advanced profiler options.

• Breaking changes:

• Deprecations:

• Bug fixes:

  • Fixes bug causing tracebacks to be swallowed in worker processes.
  • Fixes performance regression in traceback filtering when workers are frequently exhausting small datasets.
  • Fixes silent string truncation in shared memory batching.

Commits

• 7443a2a Correct next index calculation in dataloader set state.
• 50d2dd0 Fix silent string truncation in PyGrain shared memory batching
• 679e0db internal
• 21fde02 Allow Prometheus telemetry for iterator_get_next metric
• 270cb8a Update TfMixtureIndexSampler to support proportions with rank 1 or 2
• 90f21e7 bug fix
• 7057c45 Force index shuffle rounds to not be bigger than 1024.
• a85929f Add comprehensive docstrings for Grain classes (RandomAccessDataSource, Datas...
• fc7c9e7 bugfix in stats module.
• 302a288 Fixed documentation
• Additional commits viewable in compare view

Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting @dependabot rebase.

---

Dependabot commands and options

You can trigger Dependabot actions by commenting on this PR:

• @dependabot rebase will rebase this PR
• @dependabot recreate will recreate this PR, overwriting any edits that have been made to it
• @dependabot show <dependency name> ignore conditions will show all of the ignore conditions of the specified dependency
• @dependabot ignore this major version will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this minor version will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself)
• @dependabot ignore this dependency will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself)

[dependabot]

Superseded by #337.