https://github.com/adeelh/pytorch-sequential-helpers

Some helper PyTorch modules that allow complex networks to be expressed as Sequentials
https://github.com/adeelh/pytorch-sequential-helpers

deep-learning neural-network neural-networks pytorch

Last synced: about 1 month ago
JSON representation

Some helper PyTorch modules that allow complex networks to be expressed as Sequentials

• Host: GitHub
• URL: https://github.com/adeelh/pytorch-sequential-helpers
• Owner: AdeelH
• License: mit
• Created: 2019-08-18T01:02:50.000Z (almost 7 years ago)
• Default Branch: master
• Last Pushed: 2020-08-07T13:50:50.000Z (almost 6 years ago)
• Last Synced: 2025-04-11T15:09:57.157Z (about 1 year ago)
• Topics: deep-learning, neural-network, neural-networks, pytorch
• Language: Python
• Homepage:
• Size: 6.84 KB
• Stars: 2
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# pytorch-sequential-helpers

Some helper modules that allow complex networks (particulary those with parallel data flows) to be expressed as a single Sequential.

# Example

**Pass same input to 2 different NN branches and merge the results by adding**

```{python3}

nn.Sequential(

  Parallel(branch1, branch2),

  Add()

)

```

**Split the RGB channels of a batch of images, pass each to a different NN branch, and concat the results**

```{python3}

nn.Sequential(

  Split((1, 1, 1), dim=1),

  Parallel(branch1, branch2, branch3),

  Concat(dim=1)

)

```