It uses Huggingface's open-source BERT pedestal and fine-tuns Huggingface's open-source financial_phrasebank dataset for analysis of financial news sentences
Now there are two parts: training and prediction
And you can use requirements.txt to install the envs