Edit model card

sa_BERT_no_pretrain_cola

This model is a fine-tuned version of on the GLUE COLA dataset. It achieves the following results on the evaluation set:

  • Loss: 0.6180
  • Matthews Correlation: 0.0
  • Accuracy: 0.6913

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: 0.0005
  • train_batch_size: 128
  • eval_batch_size: 128
  • seed: 10
  • distributed_type: multi-GPU
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 50
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Matthews Correlation Accuracy
0.8826 1.0 67 0.6624 0.0 0.6913
0.616 2.0 134 0.6358 0.0 0.6913
0.6134 3.0 201 0.6195 0.0 0.6913
0.6139 4.0 268 0.6285 0.0 0.6913
0.6117 5.0 335 0.6180 0.0 0.6913
0.6099 6.0 402 0.6183 0.0 0.6913
0.6113 7.0 469 0.6232 0.0 0.6913
0.6135 8.0 536 0.6182 0.0 0.6913
0.6094 9.0 603 0.6221 0.0 0.6913
0.6096 10.0 670 0.6310 0.0 0.6913

Framework versions

  • Transformers 4.29.2
  • Pytorch 1.14.0a0+410ce96
  • Datasets 2.12.0
  • Tokenizers 0.13.3
Downloads last month
2

Dataset used to train gokuls/sa_BERT_no_pretrain_cola

Evaluation results