Skip to content

v0.11.0
Compare
Choose a tag to compare
@NiklasKoehneckeAA NiklasKoehneckeAA released this 29 Apr 11:37
· 328 commits to main since this release
v0.11.0
e5b3a8f
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

Breaking Changes

  • breaking_change: HuggingFaceDatasetRepository now has a parameter caching, which caches a examples of a dataset once loaded. This is True by default. This drastically reduces network traffic. For a non-breaking change, set it to False.
  • breaking_change: MultipleChunkRetrieverQa does not take insert_chunk_size-parameter but instead takes ExpandChunks-task
  • breaking_change: the issue_cassification_user_journey notebook moved to its own repository

New Features

  • feature: Llama2InstructModel to support llama-2 models in Aleph Alpha API
  • feature: Llama3InstructModel to support llama-3 models in Aleph Alpha API
  • feature: ExpandChunks-task caches chunked documents by ID
  • feature: DocumentIndexClient now supports
    - create_index
    - index_configuration
    - assign_index_to_collection
    - delete_index_from_collection
    - list_assigned_index_names
  • feature: DocumentIndexRetriever now supports index_name
  • feature: Runner.run_dataset now has a configurable number of workers via max_workers and defaults to the previous value, which is 10.
  • feature: In case a BusyError is raised during a complete the LimitedConcurrencyClient will retry until max_retry_time is reached.
  • feature: FileTracer now accepts as log_file_path both, a str and a Path

Fixes

  • refactor: rename index parameter in DocumentIndex.search() to index_name
  • fix: HuggingFaceRepository no longer is a dataset repository. This also means that HuggingFaceAggregationRepository no longer is a dataset repository.

Full Changelog: v0.10.0...v0.11.0

Assets 2
Loading