https://github.com/nearai/torchfold

Tools for PyTorch
https://github.com/nearai/torchfold

deep-learning deep-neural-networks machine-learning pytorch

Last synced: 3 days ago
JSON representation

Tools for PyTorch

• Host: GitHub
• URL: https://github.com/nearai/torchfold
• Owner: nearai
• License: apache-2.0
• Created: 2017-08-22T22:28:18.000Z (over 7 years ago)
• Default Branch: master
• Last Pushed: 2022-08-10T20:26:14.000Z (over 2 years ago)
• Last Synced: 2024-05-23T06:48:01.224Z (7 months ago)
• Topics: deep-learning, deep-neural-networks, machine-learning, pytorch
• Language: Python
• Size: 73.2 KB
• Stars: 221
• Watchers: 17
• Forks: 28
• Open Issues: 7
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• Awesome-pytorch-list-CNVersion - pytorch-tools

README

        


[![PyPi version](https://pypip.in/v/torchfold/badge.png)](https://pypi.org/project/torchfold/) [![DOI](https://zenodo.org/badge/DOI/10.5281/zenodo.1299387.svg)](https://doi.org/10.5281/zenodo.1299387)

# TorchFold

Blog post: http://near.ai/articles/2017-09-06-PyTorch-Dynamic-Batching/

Analogous to [TensorFlow Fold](https://github.com/tensorflow/fold), implements dynamic batching with super simple interface.

Replace every direct call in your computation to nn module with `f.add('function name', arguments)`.

It will construct an optimized version of computation and on `f.apply` will dynamically batch and execute the computation on given nn module.

## Installation

We recommend using pip package manager:

```

pip install torchfold

```

## Example

```

    f = torchfold.Fold()

   

    def dfs(node):

        if is_leaf(node):

            return f.add('leaf', node)

        else:

            prev = f.add('init')

            for child in children(node):

                prev = f.add('child', prev, child)

            return prev

    class Model(nn.Module):

        def __init__(self, ...):

            ...

        def leaf(self, leaf):

            ...

        def child(self, prev, child):

            ...

    res = dfs(my_tree)

    model = Model(...)

    f.apply(model, [[res]])

```

To cite this repository in publications:

    @misc{illia_polosukhin_2018_1299387,

      author       = {Illia Polosukhin and

                      Maksym Zavershynskyi},

      title        = {nearai/torchfold: v0.1.0},

      month        = jun,

      year         = 2018,

      doi          = {10.5281/zenodo.1299387},

      url          = {https://doi.org/10.5281/zenodo.1299387}

    }