Edit model card

add_BERT_no_pretrain_mrpc

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

  • Loss: 0.5912
  • Accuracy: 0.6961
  • F1: 0.7933
  • Combined Score: 0.7447

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: 4e-05
  • 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

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 Combined Score
0.6854 1.0 29 0.6711 0.6838 0.8122 0.7480
0.6496 2.0 58 0.6802 0.6838 0.8122 0.7480
0.648 3.0 87 0.6246 0.6838 0.8122 0.7480
0.6363 4.0 116 0.6174 0.6838 0.8122 0.7480
0.6049 5.0 145 0.6176 0.6593 0.7459 0.7026
0.5491 6.0 174 0.6038 0.6814 0.7950 0.7382
0.5601 7.0 203 0.5912 0.6961 0.7933 0.7447
0.5505 8.0 232 0.6346 0.6716 0.7781 0.7249
0.5327 9.0 261 0.6283 0.6544 0.7531 0.7037
0.529 10.0 290 0.6341 0.6520 0.7568 0.7044
0.5337 11.0 319 0.6285 0.6618 0.7579 0.7098
0.5383 12.0 348 0.6322 0.6348 0.7286 0.6817

Framework versions

  • Transformers 4.30.2
  • Pytorch 1.14.0a0+410ce96
  • Datasets 2.12.0
  • Tokenizers 0.13.3
Downloads last month
10
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.

Dataset used to train gokuls/add_BERT_no_pretrain_mrpc

Evaluation results