Edit model card

canine-s-finetuned-cola

This model is a fine-tuned version of google/canine-s on the glue dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6653
  • Matthews Correlation: 0.0594

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: 5

Training results

Training Loss Epoch Step Validation Loss Matthews Correlation
0.6132 1.0 535 0.6289 0.0
0.6062 2.0 1070 0.6179 0.0
0.6122 3.0 1605 0.6160 0.0
0.5939 4.0 2140 0.6159 0.0
0.5721 5.0 2675 0.6653 0.0594

Framework versions

  • Transformers 4.17.0
  • Pytorch 1.10.0+cu111
  • Datasets 2.0.0
  • Tokenizers 0.11.6
Downloads last month
52

Dataset used to train vicl/canine-s-finetuned-cola

Evaluation results