https://github.com/automan000/Convolutional_LSTM_PyTorch

Multi-layer convolutional LSTM with Pytorch
https://github.com/automan000/Convolutional_LSTM_PyTorch

Last synced: 2 months ago
JSON representation

Multi-layer convolutional LSTM with Pytorch

• Host: GitHub
• URL: https://github.com/automan000/Convolutional_LSTM_PyTorch
• Owner: automan000
• Archived: true
• Created: 2017-06-07T11:45:29.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2020-09-27T06:50:12.000Z (over 4 years ago)
• Last Synced: 2024-05-23T06:46:43.616Z (8 months ago)
• Language: Python
• Homepage:
• Size: 7.81 KB
• Stars: 785
• Watchers: 6
• Forks: 199
• Open Issues: 10
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-pytorch-list-CNVersion - Convolution_LSTM_pytorch

README

        
# Convolution_LSTM_pytorch

Thanks for your attention. 

I haven't got time to maintain this repo for a long time.

I recommend this [repo](https://github.com/Hzzone/Precipitation-Nowcasting) which provides an excellent implementation.

# Usage

A multi-layer convolution LSTM module

Pytorch implementation of  [Convolutional LSTM Network: A Machine Learning Approach for Precipitation Nowcasting](https://arxiv.org/abs/1506.04214)

```python

clstm = ConvLSTM(input_channels=512, hidden_channels=[128, 64, 64], kernel_size=5, step=9, effective_step=[2, 4, 8])

lstm_outputs = clstm(cnn_features)

hidden_states = lstm_outputs[0]

```

# Thanks

Thanks to [@Jackie-Chou](https://github.com/Jackie-Chou) and [@chencodeX](https://github.com/chencodeX) who provide lots of valuable advice.

I apology for the inconvenience.