Changelog¶
All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
[0.1.11] - 2021-01-18¶
Changed¶
- Adding rank feature to serving parse fn by default and removing dependence on required serving_info attribute
[0.1.9] - 2021-11-29¶
Changed¶
- Refactored secondary label metrics computation for ranking and added unit tests
- Added NDCG metric for secondary labels
[0.1.5] - 2021-07-15¶
Added¶
- Adding support for performing post-training steps (such as copying data) by custom class inheriting RelevancePipeline.
[0.1.4] - 2021-06-30¶
Changed¶
- Performing pre-processing step in
__init__()to be able to copy files before model_config and feature_config are initiated.
[0.1.2] - 2021-06-16¶
Added¶
- Support for performing pre-processing steps (such as copying data) by custom class inheriting RelevancePipeline.
[0.1.1] - 2021-05-20¶
Added¶
- Support for using native tf/keras feature functions from the feature config YAML
[0.1.0] - 2021-03-01¶
Changed¶
- TFRecord format changed for SequenceExample to earlier implementation.
- Removed support for
max_lenattribute for SequenceExample features. - No effective changes for Example TFRecords.
- TFRecord implementation on python (training) and jvm (inference) side are now in sync.
[0.0.5] - 2021-02-17¶
Added¶
- Changelog file to track version updates for ml4ir.
build-requirements.txtwith all python dependencies needed for developing on ml4ir and the CircleCI autobuilds.- Updated CircleCI builds to use
build-requirements.txt
Fixed¶
- Removed build requirements from the base ml4ir
requirements.txtallowing us to keep the published whl file dependencies to be minimal.