Edit model card

nateraw/bert-base-uncased-emotion converted to ONNX and quantized using optimum.


Model description

bert-base-uncased finetuned on the emotion dataset using PyTorch Lightning. Sequence length 128, learning rate 2e-5, batch size 32, 2 GPUs, 4 epochs.

For more details, please see, the emotion dataset on nlp viewer.

Limitations and bias

  • Not the best model, but it works in a pinch I guess...
  • Code not available as I just hacked this together.
  • Follow me on github to get notified when code is made available.

Training data

Data came from HuggingFace's datasets package. The data can be viewed on nlp viewer.

Training procedure


Eval results

val_acc - 0.931 (useless, as this should be precision/recall/f1)

The score was calculated using PyTorch Lightning metrics.

Downloads last month

Dataset used to train Cohee/bert-base-uncased-emotion-onnx