v2.0.3

Release date 21/03/2022

• #384: Unpin tokenizers to allow users to upgrade to latest transformers

v2.0.2

Merge pull request #379 from wellcometrust/fix/move-flake8-to-test-setup

Move flake8 and black to tests

v2.0.1

Merge pull request #374 from wellcometrust/release-version-2.0.0

Release 2.0.0

v2.0.0

Improvements:

• #328: Expose build_model to CNNClassifier and don't rebuild in fit
• #372: Add more fine grained extras to make vanilla WellcomeML light (109MB)
• #368: Expand LOGGING_LEVEL to TF_CPP_MIN_LOG_LEVEL
• #332: Add wellcomeml.viz to vizualize clusters
• #344: Add filter by variable in visualize clusters

Breaking changes:

• #371: Delete wellcomeml.ml.__init__ so all ml models need to be explicitly imported

Bug fixes:

• #357: Fix sent2vec test error

v1.2.1

Merge pull request #346 from wellcometrust/fix/tf-idf-fit-transform

Fix/tf idf fit transform

v1.2.0

Improvements:

#327: Adds verbose and tensorboard_log_path to CNN and SemanticEquivalenceClassifier
#315: Implements decode to KerasTokenizer and TransformersTokenizer.
#308: Adds EPMCClient to download data from EPMC
#283: Disable umap to make imports in wellcomeml faster
#279: Extend LOGGING_LEVEL env variable to control more libraries logger
#306: Break down deep-learning extra to tensorflow,torch,spacy for more control over what's installed
#289: Re-factors frequency vectoriser saving function to make it more efficient
Bug fixes:

#313: Fix concat feature_approach in CNN
#297: Fix OOM error in CNN predict
#292: Fix clustering pipeline when umap used and input length > 4096

v1.1.0

Merge pull request #276 from wellcometrust/feature/clustering-improve…

…ments

Feature/clustering improvements

v1.0.2

Reduce scope and improve reproducibility

v1.0.1

Add warning about early stopping not working

v2021.2.1

Bug fixes

• Update version to 2021.2.1 which was out of sync in latest release #206