YAML Metadata Warning:empty or missing yaml metadata in repo card

Check out the documentation for more information.

EmoSense BERT Text Model

This folder is the default output target for:

python scripts/train_bert_text_model.py

After training, it should contain Hugging Face model artifacts such as:

  • config.json
  • model.safetensors or pytorch_model.bin
  • tokenizer files
  • training_args.bin
  • emosense_text_model_config.json

The classifier predicts these EmoSense text-channel labels:

  • Anger
  • Contempt
  • Disgust
  • Enjoyment
  • Fear
  • Sadness
  • Surprise

Use the model softmax output as the text channel before EmoSense applies late fusion with the facial channel.

Downloads last month
10
Safetensors
Model size
0.1B params
Tensor type
F32
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Space using jeanryy/emosense-bert-model 1