xnli_xlm_r_only_bg / README.md
Dan Semin
update model card README.md
fd4fd4b
metadata
license: mit
tags:
  - text-classification
  - generated_from_trainer
datasets:
  - xnli
metrics:
  - accuracy
model-index:
  - name: xnli_xlm_r_only_bg
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: xnli
          type: xnli
          config: bg
          split: train
          args: bg
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.7839357429718875

xnli_xlm_r_only_bg

This model is a fine-tuned version of xlm-roberta-base on the xnli dataset. It achieves the following results on the evaluation set:

  • Loss: 0.7896
  • Accuracy: 0.7839

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: 2e-05
  • train_batch_size: 128
  • eval_batch_size: 128
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 10

Training results

Training Loss Epoch Step Validation Loss Accuracy
0.6649 1.0 3068 0.5678 0.7659
0.5321 2.0 6136 0.5338 0.7932
0.4668 3.0 9204 0.5648 0.7871
0.4129 4.0 12272 0.5736 0.7835
0.365 5.0 15340 0.5782 0.7964
0.3202 6.0 18408 0.6482 0.7847
0.2842 7.0 21476 0.6565 0.7900
0.2533 8.0 24544 0.7211 0.7912
0.2278 9.0 27612 0.7751 0.7815
0.2102 10.0 30680 0.7896 0.7839

Framework versions

  • Transformers 4.24.0
  • Pytorch 1.13.0
  • Datasets 2.6.1
  • Tokenizers 0.13.1