https://github.com/nitinprakash96/nmt

https://github.com/nitinprakash96/nmt

deep-learning machine-translation neural-network pytorch

Last synced: 9 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/nitinprakash96/nmt
• Owner: nitinprakash96
• Created: 2019-03-28T07:05:56.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2019-04-01T13:39:26.000Z (almost 6 years ago)
• Last Synced: 2025-01-07T23:36:16.129Z (18 days ago)
• Topics: deep-learning, machine-translation, neural-network, pytorch
• Language: Python
• Size: 14 MB
• Stars: 1
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
## Neural Machine Translation

This is an attempt to learn the implementation of NMT model and achieve a BLEU score larger than 21.

### File Structure

- model.py: File containing the model structure.

- vocab.py: Script used to generate `vocab.json` vocabulary file.

- util.py: File containing helper functions.

- run.py: File to facilitate training and evaluation of the model.

- embeddings.py: Model embeddings

### Usage

- To generate vocabulary

    `sh run.sh vocab` or `python vocab.py --train-src=./data/train.es --train-tgt=./data/train.en vocab.json`

- To train the model

    `python run.py train --train-src=./data/train.es --train-tgt=./data/train.en --dev-src=./data/dev.es --dev-tgt=./data/dev.en --vocab=vocab.json`

- To test the model

    `python run.py decode model.bin ./data/test.es ./data/test.en outputs/test_outputs.txt`