aayux

L2 Regularization for all layers

6

It is better practice to apply weight decay methods in each layer. This commit proposes to change the current implementation of L2 Regularization from just the output layer to the...

__future__ imports for Python 3 compatibilty

1

- Python 3 style division to resolve bug in #24. - Other \_\_future\_\_ imports for Python 3 compatibility. Finally, this could also break intended behaviour in some places so: -...

Looking for contributors to help with major code cleanup and refactoring based on PEP-8 style guidelines.

Self attention for pooling linear classifier

5

Add a `BiAttentionPoolingClassifier` (self attention for pooling linear classifier) as in [Attention is all you need](https://arxiv.org/abs/1706.03762) following the discussion with @sebastianruder in Teams. I ran out of memory on my...