Edit model card

yes_no_qna_deberta_model

This model is a fine-tuned version of microsoft/deberta-v3-base on the super_glue dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5570
  • Accuracy: 0.8508

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 2e-05
  • train_batch_size: 16
  • eval_batch_size: 16
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.583 1.0 590 0.4086 0.8251
0.348 2.0 1180 0.4170 0.8465
0.2183 3.0 1770 0.5570 0.8508

Framework versions

  • Transformers 4.25.1
  • Pytorch 1.13.0+cu116
  • Datasets 2.8.0
  • Tokenizers 0.13.2
Downloads last month
150
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Finetuned from

Dataset used to train nc33/yes_no_qna_deberta_model

Evaluation results