Changelog

Version 0.9.0

Finished core features:
- Attention Mechanism
- Diffusion Graph Convolution
- GRU based on the above
- reimplementation of the model used in https://milets19.github.io/papers/milets19_paper_8.pdf

Version 0.9.1

Wrote README.md
Added requirements
uploading to TestPyPi

Version 0.9.2

fixing README.md
removed pyscaffold spam from setup.cfg
uploading to TestPyPi and testing if it installs correctly

Version 0.9.3

Final upload to real pypi and final test

Version 1.0.0

All good. Ready for release.

Version 1.0.1

fixing naming bug where we could not use more than one AttentionMechanism due to all such mechanisms having the same name but names need to be unique.

Version 1.0.2

Added RaggedTensorFeeder utensil that is able to feed a ragged tensor as

data to an RNN model that can only train on non ragged batches.

Version 1.0.3

Fixed using an average layer for only one attention head.
Fixed issue with many weights not being trained due to kwargs being passed to the parent object in the layer definitions.