Skip to content

georgeyiasemis/2D-Convolutional-Recurrent-Neural-Networks-with-PyTorch

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 

Repository files navigation

2D Convolutional Recurrent Neural Networks with PyTorch

Two dimensional Convolutional Recurrent Neural Networks implemented in PyTorch

The architecture of Conv2dLSTMCell was inspired by "Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting" (https://arxiv.org/pdf/1506.04214.pdf).

See the image below for the key equations of Conv2dLSTMCell:

Capture

The implementations of Conv2dRNNCell and Conv2dGRUCell are based on the implementation of Convolutional LSTM.

This repo contains implementations of:

  • Conv2dRNNCell
  • Conv2dLSTMCell
  • Conv2dGRUCell

and

  • Conv2dRNN / Biderectional Conv2dRNN
  • Conv2dLSTM / Biderectional Conv2dLSTM
  • Conv2dGRU / Biderectional Conv2dGRU.

Dependencies

  • pytorch
  • numpy

Releases

No releases published

Packages

No packages published

Languages