Data augmenation in Natural Language Processing (NLP) for text classification. Techniques: BERT, Fine-tune BERT ----- update in 3rd June ------ add new DA techs, which are Back translation, Fasttext and Words deletion. ----- update in 5th June ------ Tuned BERT classification for all tasks ----- update 7th June -------- Try use other dataset to test the effect of data augmentation
Roger-G/Data-augmentation
this is a data augmentation project to solve data imbalance problem
Jupyter Notebook