idimohammed/HausaBERTa

HausaBERTa is a pretrained lexicon low resources language model. The model was trained on Hausa Language. You can use this model with Transformers for sentiment analysis task in Hausa Language. # HausaBERTa HausaBERTa is a pretrained lexicon low resources language model. The model was trained on Hausa Language. You can use this model with Transformers for sentiment analysis task in Hausa Language. The comprised folder name "Hausa Model". This folder consist of 3-sub folders. The "Fine Tune" folder consist of 3-files "hausaBERTa_fine_tuning.ipynb", "hausaBERTa_fine_tuning.py", and "hausaBERTdatatrain.txt". These files are used for Fine tuning BERT Model. Depends on your environment, "hausaBERTa_fine_tuning.ipynb" and "hausaBERTa_fine_tuning.py" do thesame function while "hausaBERTdatatrain.txt" is the training dataset. The hausaBERTa_fine_tuning files consist of instructions on how to fine tune your BERT model if you wish to train with our dataset on your dataset. The Comparison folder consist of six files which provide a comparison of HausaBERTa with Vader, TextBlob, Flair, and AfriBERTa models. The source codes is labeled with description for user to understand. The Runing Model folder contain program writen in Python that utilized HausaBERTa model for sentiment analysis in one sentence or bulk file.