gokuls's picture
End of training
970dabe
metadata
license: apache-2.0
base_model: google/bert_uncased_L-6_H-512_A-8
tags:
  - generated_from_trainer
datasets:
  - emotion
metrics:
  - accuracy
model-index:
  - name: bert_uncased_L-6_H-512_A-8_emotion
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: emotion
          type: emotion
          config: split
          split: validation
          args: split
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.941

bert_uncased_L-6_H-512_A-8_emotion

This model is a fine-tuned version of google/bert_uncased_L-6_H-512_A-8 on the emotion dataset. It achieves the following results on the evaluation set:

  • Loss: 0.1465
  • Accuracy: 0.941

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: 5e-05
  • train_batch_size: 64
  • eval_batch_size: 64
  • seed: 33
  • distributed_type: multi-GPU
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6249 1.0 250 0.1965 0.9275
0.1782 2.0 500 0.1533 0.938
0.1245 3.0 750 0.1467 0.9365
0.0951 4.0 1000 0.1480 0.94
0.0764 5.0 1250 0.1465 0.941
0.0634 6.0 1500 0.1594 0.94
0.0422 7.0 1750 0.2059 0.935
0.0381 8.0 2000 0.1881 0.938
0.027 9.0 2250 0.2025 0.9405
0.0203 10.0 2500 0.2032 0.9375

Framework versions

  • Transformers 4.34.0
  • Pytorch 1.14.0a0+410ce96
  • Datasets 2.14.5
  • Tokenizers 0.14.1