1.3.1

• Update to OK Bloomer 1.0 stable

1.3.0

• Switch back to the original fork of Tensor
• Added maxBins hyper-parameter to CART-based learners
• Added stream Deduplicator extractor
• Added the SiLU activation function
• Added Swish activation layer

1.2.3

• Fix Multiclass layer cross-entropy gradient optimization

1.2.2

• Allow empty dataset objects in stack()

1.2.1

• Refactor stratified methods on Labeled dataset
• Narrower typehints

1.2.0

• Added Logit Boost classifier
• Interval Discretizer variable or equi-depth binning
• Text Normalizers now lower or upper case

1.1.3

• Min Max Normalizer compensate for 0 variance features

1.1.2

• Improved random floating point number precision
• Deduplicate Preset seeder centroids
• Fix Gradient Boost learning rate upper bound
• Fix Loda histogram edge alignment

1.1.1

• Fix Gradient Boost subsampling and importance scores

1.1.0

• Update to Scienide Tensor 3.0
• Added Nesterov's lookahead to Momentum Optimizer
• Added Reversible transformer interface
• MaxAbs, Z Score, and Robust scalers are now Reversible
• Min Max Normalizer now implements Reversible
• TF-IDF Transformer is now Reversible
• Added Preset cluster seeder
• Added Concatenator extractor