Torch Containers simplified in PyTorch
This repository aims to help former Torchies more seamlessly transition to the “Containerless” world of PyTorch by providing a list of PyTorch implementations of Torch Table Layers.
Note: As a result of full integration with autograd, PyTorch requires networks to be defined in the following manner:
- Define all layers to be used in the
__init__method of your network - Combine them however you want in the
forwardmethod of your network (avoiding in place Tensor ops)
And that’s all there is to it!
We will build upon a generic “TableModule” class that we initially define as: