NLP for Oriya
August 7, 2020 ยท View on GitHub
This repository contains State of the Art Language models and Classifier for Oriya, which is spoken in the Indian state of Odisha.
The models trained here have been used in Natural Language Toolkit for Indic Languages (iNLTK)
Dataset
Created as part of this project
Open Source Datasets
Results
Language Model Perplexity (on validation set)
| Architecture/Dataset | Oriya Wikipedia Articles |
|---|---|
| ULMFiT | 26.57 |
| TransformerXL | 26.81 |
Classification Metrics
ULMFiT
| Dataset | Accuracy | MCC | Notebook to Reproduce results |
|---|---|---|---|
| IndicNLP News Article Classification Dataset - Oriya | 98.83 | 98.44 | Link |
Visualizations
Word Embeddings
| Architecture | Visualization |
|---|---|
| ULMFiT | Embeddings projection |
| TransformerXL | Embeddings projection |
Pretrained Models
Language Models
Download pretrained Language Models from here
Tokenizer
Trained tokenizer using Google's sentencepiece
Download the trained model and vocabulary from here