gokuls's picture
End of training
742c7a4
metadata
language:
  - en
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - spearmanr
model-index:
  - name: hBERTv2_new_pretrain_w_init_48_stsb
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE STSB
          type: glue
          config: stsb
          split: validation
          args: stsb
        metrics:
          - name: Spearmanr
            type: spearmanr
            value: 0.25026626408652064

hBERTv2_new_pretrain_w_init_48_stsb

This model is a fine-tuned version of gokuls/bert_12_layer_model_v2_complete_training_new_wt_init_48 on the GLUE STSB dataset. It achieves the following results on the evaluation set:

  • Loss: 2.2569
  • Pearson: 0.2426
  • Spearmanr: 0.2503
  • Combined Score: 0.2465

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 Pearson Spearmanr Combined Score
2.3064 1.0 45 2.2784 0.1492 0.1433 0.1462
1.9454 2.0 90 2.2773 0.2660 0.2546 0.2603
1.7733 3.0 135 2.2569 0.2426 0.2503 0.2465
1.4752 4.0 180 2.3395 0.2833 0.3025 0.2929
1.083 5.0 225 2.4140 0.3017 0.3066 0.3042
0.8694 6.0 270 2.8854 0.2790 0.3025 0.2908
0.7272 7.0 315 2.8901 0.3041 0.3066 0.3054
0.6008 8.0 360 2.6252 0.2837 0.2833 0.2835

Framework versions

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