https://github.com/mast-group/convolutional-attention

Repository for the code of the "A Convolutional Attention Network for Extreme Summarization of Source Code" paper
https://github.com/mast-group/convolutional-attention

convolutional-attention-network machine-learning ml4code

Last synced: 12 days ago
JSON representation

Repository for the code of the "A Convolutional Attention Network for Extreme Summarization of Source Code" paper

• Host: GitHub
• URL: https://github.com/mast-group/convolutional-attention
• Owner: mast-group
• License: bsd-3-clause
• Archived: true
• Created: 2016-05-07T00:20:14.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2016-07-19T09:27:39.000Z (over 8 years ago)
• Last Synced: 2024-07-31T07:15:24.222Z (3 months ago)
• Topics: convolutional-attention-network, machine-learning, ml4code
• Language: HTML
• Homepage:
• Size: 1.59 MB
• Stars: 119
• Watchers: 11
• Forks: 31
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• awesome-summ - [code

README

        
Convolutional Attention Network

===============

Code related to the paper:

```

@inproceedings{allamanis2016convolutional,

  title={A Convolutional Attention Network for Extreme Summarization of Source Code},

  author={Allamanis, Miltiadis and Peng, Hao and Sutton, Charles},

  booktitle={International Conference on Machine Learning (ICML)},

  year={2016}

}

```

For more information and the data of the paper, see [here](http://groups.inf.ed.ac.uk/cup/codeattention/).

The project depends on Theano and uses Python 2.7.

Usage Instructions

======

To train the `copy_attention` model with the data use

```

> python copy_conv_rec_learner.py    

```

were `D` is the embedding space dimenssion (128 in paper.)

The best model will be saved at `.pkl`

To evaluate an existing model re-run with exactly the same parameteres except

for `` which should be zero.

The following code will generate names from a pre-trained model and a test_file

with code examples.

```python

model = ConvolutionalCopyAttentionalRecurrentLearner.load(model_fname)

test_data, original_names = model.naming_data.get_data_in_recurrent_copy_convolution_format(test_file, model.padding_size)

test_name_targets, test_code_sentences, test_code, test_target_is_unk, test_copy_vectors = test_data

idx = 2  # pick an example from test_file

res = model.predict_name(np.atleast_2d(test_code[idx]))

print "original name:", ' '.join(original_names[idx].split(','))

print "code:", ' '.join(test_code[idx])

print "generated names:"

for r,v in res:

    print v, ' '.join(r)

```