gokuls's picture
End of training
f49c88e
metadata
language:
  - en
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
  - f1
model-index:
  - name: hBERTv1_new_pretrain_w_init_48_qqp
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE QQP
          type: glue
          config: qqp
          split: validation
          args: qqp
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.8430373485035865
          - name: F1
            type: f1
            value: 0.7845307619176966

hBERTv1_new_pretrain_w_init_48_qqp

This model is a fine-tuned version of gokuls/bert_12_layer_model_v1_complete_training_new_wt_init_48 on the GLUE QQP dataset. It achieves the following results on the evaluation set:

  • Loss: 0.3476
  • Accuracy: 0.8430
  • F1: 0.7845
  • Combined Score: 0.8138

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
0.4637 1.0 2843 0.3907 0.8136 0.7636 0.7886
0.363 2.0 5686 0.3536 0.8338 0.7900 0.8119
0.3211 3.0 8529 0.3476 0.8430 0.7845 0.8138
0.2906 4.0 11372 0.3539 0.8531 0.8059 0.8295
0.2603 5.0 14215 0.3531 0.8531 0.8017 0.8274
0.2373 6.0 17058 0.3716 0.8561 0.8089 0.8325
0.2175 7.0 19901 0.3553 0.8565 0.8123 0.8344
0.1957 8.0 22744 0.3726 0.8551 0.8099 0.8325

Framework versions

  • Transformers 4.29.2
  • Pytorch 1.14.0a0+410ce96
  • Datasets 2.12.0
  • Tokenizers 0.13.3