YAML Metadata Warning:empty or missing yaml metadata in repo card
Check out the documentation for more information.
Inference of emotions in finance context
this is a simple project created as part of a learning experience on the intricate functioning of LLM's and transformers. it uses a BERT (encoder-only) model to infer emotions from small phrases in finance. emotions can be classified as either one of seven labels: neutral, happy, fear, anger, sad, disgust, surprise.
feats
uses BERT (encoder-only) for sequence classification.
tokenization with bert-base-cased.
custom mapping from emotion labels to numeric IDs.
training and evaluation with Hugging Face Trainer.
accuracy metric using evaluate.
- Downloads last month
- 5
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support