https://github.com/sagorbrur/bntranslit

Bangla Transliteration Package
https://github.com/sagorbrur/bntranslit

bangla bangla-transliteration bengali bengali-transliteration deep-learning nlp pytorch transliteration

Last synced: 4 days ago
JSON representation

Bangla Transliteration Package

• Host: GitHub
• URL: https://github.com/sagorbrur/bntranslit
• Owner: sagorbrur
• License: mit
• Created: 2021-04-28T16:52:35.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2021-05-05T14:03:27.000Z (over 3 years ago)
• Last Synced: 2024-12-17T03:34:50.406Z (17 days ago)
• Topics: bangla, bangla-transliteration, bengali, bengali-transliteration, deep-learning, nlp, pytorch, transliteration
• Language: Python
• Homepage:
• Size: 35.3 MB
• Stars: 8
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: license

Awesome Lists containing this project

README

        
# BNTRANSLIT

__BNTRANSLIT__ is a deep learning based transliteration app for Bangla word.

## Installation

`pip install bntranslit`

## Dependency

- pytorch 1.7.0 or 1.7.0+

NB: No `GPU` Needed. Totally `CPU` based

## Pre-trained Model

- [Download bntranslit_model](https://github.com/sagorbrur/bntranslit/raw/master/model/bntranslit_model.pth)

## Usage

```py

from bntranslit import BNTransliteration

model_path = "bntranslit_model.pth"

bntrans = BNTransliteration(model_path)

word = "aami"

output = bntrans.predict(word, topk=10)

# output: ['আমি', 'আমী', 'অ্যামি', 'আমিই', 'এমি', 'আমির', 'আমিদ', 'আমই', 'আমে', 'আমিতে']

```

## Datasets and Training Details

- We used [Google Dakshina Dataset](https://github.com/google-research-datasets/dakshina)

- Thanks to [AI4Bharat](https://github.com/AI4Bharat/IndianNLP-Transliteration) for providing training notebook with details explanation

- We trained Google Bangla Dakshina lexicons train datasets for 10 epochs with batch size 128, 1e-3, embedding dim = 300, hidden dim = 512, lstm, used attention

- We evaluated our trained model with Google Bangla Dakshina lexicon test data using [AI4Bharat evaluation script](https://raw.githubusercontent.com/AI4Bharat/IndianNLP-Transliteration/jgeob-dev/tools/accuracy_reporter/accuracy_news.py) and our evaluation results insides `docs/evaluation_summary.txt`