https://github.com/pavanchhatpar/copynet-tf

CopyNet (Copy Mechanism in Seq2Seq) implementation with TensorFlow 2
https://github.com/pavanchhatpar/copynet-tf

copy-mechanism copynet seq2seq tensorflow-tutorial tensorflow2 tf-keras

Last synced: 3 months ago
JSON representation

CopyNet (Copy Mechanism in Seq2Seq) implementation with TensorFlow 2

• Host: GitHub
• URL: https://github.com/pavanchhatpar/copynet-tf
• Owner: pavanchhatpar
• Created: 2020-04-30T11:52:18.000Z (over 4 years ago)
• Default Branch: master
• Last Pushed: 2022-11-21T21:18:49.000Z (about 2 years ago)
• Last Synced: 2024-09-27T21:21:52.810Z (3 months ago)
• Topics: copy-mechanism, copynet, seq2seq, tensorflow-tutorial, tensorflow2, tf-keras
• Language: Python
• Size: 105 KB
• Stars: 10
• Watchers: 3
• Forks: 0
• Open Issues: 4
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# CopyNet implementation with TensorFlow 2

 - Incorporating Copying Mechanism in Sequence-to-Sequence Learning

 - Uses `TensorFlow 2.0` and above APIs with `tf.keras` too

 - Adapted from AllenNLP's PyTorch implementation, their blog referenced 

 below was very helpful to understand the math from an implementation

 perspective

![Python package](https://github.com/pavanchhatpar/copynet-tf/workflows/Python%20package/badge.svg)

![Upload Python Package](https://github.com/pavanchhatpar/copynet-tf/workflows/Upload%20Python%20Package/badge.svg)

## Install package

### Using pip

```bash

pip install copynet-tf

```

### Compile from source

```bash

python -m pip install --upgrade pip

pip install setuptools wheel

python setup.py sdist bdist_wheel

```

## Examples

### Abstract usage

```python

...

import MyEncoder

from copynet_tf import GRUDecoder

...

class MyModel(tf.keras.Model):

  ...

  def call(self, X, y, training):

    source_token_ids, source2target_ids = X

    target_token_ids, target2source_ids = y

    enc_output, enc_final_output, mask = self.encoder(X, y, training)

    output_dict = self.decoder(

      source_token_ids, source2target_ids, mask,

      target_token_ids, target2source_ids, training)

    return output_dict

  ...

```

### Concrete examples

Find concrete examples inside [examples](./examples) folder of the repo

## References

 - Incorporating Copying Mechanism in Sequence-to-Sequence Learning: ([paper](https://arxiv.org/abs/1603.06393))

 - AllenNLP implementation: ([blog](https://medium.com/@epwalsh10/incorporating-a-copy-mechanism-into-sequence-to-sequence-models-40917280b89d)) ([code](https://github.com/epwalsh/nlp-models))

 - BLEU score metric: ([code](https://github.com/tensorflow/nmt/blob/master/nmt/scripts/bleu.py))