David Mascharka

Refactor mygrad.nnet.layers

3

Opening this as a draft so I can get some comments. This changes the overall structure of the `mygrad.nnet.layers` module. The changes are: * Everything that was in the top-level...

`imresize` was removed in scipy 1.3

2

deprecated in 1.0; they suggest using straight PIL instead

Merge in all the text from https://github.com/CogWorksBWSI/detection to kill sync issues

Create basic ViT model

2

Add a basic vision transformer (ViT) model for classification. Sample definition for training on MNIST: ```python >>> model = ViT( ... input_shape=(28, 28), ... input_dim=1, ... patch_shape=(4, 4), ... feature_dim=128,...

Add support for exporting models

1

The current solution looks like iterating through the model parameters and `np.save`ing all the weights, then `np.load`ing them in a new model. We should support actually serializing a model

to finish: add some text about what we're doing, ~wait on the `repeat` PR from MyGrad to be merged~ Closes #49

Again could easily do an MNIST or CIFAR example with some neat visualizations. Need to do #47 first.

Easy enough to put together for MNIST once [MyGrad#169](https://github.com/rsokl/MyGrad/issues/169) is added

For two tensors

The paper is available [here](https://openreview.net/forum?id=ryQu7f-RZ).