Hartunka's picture
End of training
afc95e8 verified
metadata
library_name: transformers
language:
  - en
base_model: Hartunka/bert_base_km_5_v2
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
model-index:
  - name: bert_base_km_5_v2_wnli
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: GLUE WNLI
          type: glue
          args: wnli
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.2676056338028169

bert_base_km_5_v2_wnli

This model is a fine-tuned version of Hartunka/bert_base_km_5_v2 on the GLUE WNLI dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7557
  • Accuracy: 0.2676

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: 256
  • eval_batch_size: 256
  • seed: 10
  • optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
  • lr_scheduler_type: linear
  • num_epochs: 50

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.7687 1.0 3 0.7557 0.2676
0.7017 2.0 6 0.7759 0.2113
0.6899 3.0 9 0.7978 0.1690
0.6828 4.0 12 0.8442 0.1549
0.6741 5.0 15 0.8997 0.1549
0.6788 6.0 18 0.9396 0.1549

Framework versions

  • Transformers 4.50.2
  • Pytorch 2.2.1+cu121
  • Datasets 2.18.0
  • Tokenizers 0.21.1