TaMillion
This is a first attempt at a Tamil language model trained with Google Research's ELECTRA.
Tokenization and pre-training CoLab: https://colab.research.google.com/drive/1GngBFn_Ge5Hd2XI2febBhZyU7GDiqw5w
V2 (current): 190,000 steps
V1: 100,000 steps;
Usage
Sudalai Rajkumar's Tamil-NLP page contains classification and regression tasks: https://www.kaggle.com/sudalairajkumar/tamil-nlp
Notebook: https://colab.research.google.com/drive/1_rW9HZb6G87-5DraxHvhPOzGmSMUc67_?usp=sharin
The model outperformed mBERT on news classification: (Random: 16.7%, mBERT: 53.0%, TaMillion: 68.2%)
The model slightly outperformed mBERT on movie reviews: (RMSE - mBERT: 0.657, TaMillion: 0.626)
Equivalent accuracy on the Tirukkural topic task.
Corpus
Trained on a web crawl from https://oscar-corpus.com/ (deduped version, 5.1GB) and 1 July 2020 dump of ta.wikipedia.org (476MB)
Vocabulary
Included as vocab.txt in the upload - vocab_size is 40161