Metadata-Version: 2.1
Name: TorchCRF
Version: 1.0.5
Summary: An Implementation of Conditional Random Fields in pytorch
Home-page: https://github.com/s14t284/TorchCRF
Author: Ryuya Ikeda
Author-email: rikeda71@gmail.com
License: MIT
Keywords: crf,conditional random fields,nlp,natural language processing
Platform: UNKNOWN
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Topic :: Text Processing
Description-Content-Type: text/markdown
Requires-Dist: numpy
Requires-Dist: torch (>=1.0.0)

# Torch CRF

[![CircleCI](https://circleci.com/gh/s14t284/TorchCRF.svg?style=svg)](https://circleci.com/gh/s14t284/TorchCRF) [![Coverage Status](https://coveralls.io/repos/github/s14t284/TorchCRF/badge.svg)](https://coveralls.io/github/s14t284/TorchCRF)

Implementation of CRF (Conditional Random Fields) in PyTorch

## Requirements

- python3 (>=3.6)
- PyTorch (>=1.0)

## Installation

    `$ pip install TorchCRF`

## Usage

    ```python
    >>> import torch
    >>> from TorchCRF import CRF
    >>> device = "cuda" if torch.cuda.is_available() else "cpu"
    >>> batch_size = 2
    >>> sequence_size = 3
    >>> num_labels = 5
    >>> mask = torch.ByteTensor([[1, 1, 1], [1, 1, 0]]).to(device) # (batch_size. sequence_size)
    >>> labels = torch.LongTensor([[0, 2, 3], [1, 4, 1]]).to(device)  # (batch_size, sequence_size)
    >>> hidden = torch.randn((batch_size, sequence_size, num_labels), requires_grad=True).to(device)
    >>> crf = CRF(num_labels)
    ```

### Computing log-likelihood (used where forward)

    ```python
    >>> crf.forward(hidden, labels, mask)
    tensor([-7.6204, -3.6124], device='cuda:0', grad_fn=<ThSubBackward>)
    ```

### Decoding (predict labels of sequences)

    ```python
    >>> crf.viterbi_decode(hidden, mask)
    [[0, 2, 2], [4, 0]]
    ```

## License

MIT

## References

- [threelittlemonkeys/lstm-crf-pytorch](https://github.com/threelittlemonkeys/lstm-crf-pytorch)
- [kmkurn/pytorch-crf](https://github.com/kmkurn/pytorch-crf)


