nbhimte's picture
End of training
00a8e7e
|
raw
history blame
2.14 kB
metadata
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
model-index:
  - name: tiny-bert-mnli-distilled
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: glue
          type: glue
          args: mnli
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.5818644931227712

tiny-bert-mnli-distilled

This model is a fine-tuned version of M-FAC/bert-mini-finetuned-mnli on the glue dataset. It achieves the following results on the evaluation set:

  • Loss: 1.5018
  • Accuracy: 0.5819
  • F1 score: 0.5782
  • Precision score: 0.6036
  • Metric recall: 0.5819

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: 0.0005
  • train_batch_size: 64
  • eval_batch_size: 32
  • seed: 33
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 4
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy F1 score Precision score Metric recall
1.4475 1.0 614 1.4296 0.4521 0.4070 0.5621 0.4521
1.3354 2.0 1228 1.4320 0.4805 0.4579 0.5276 0.4805
1.2244 3.0 1842 1.4786 0.5699 0.5602 0.5865 0.5699
1.1416 4.0 2456 1.5018 0.5819 0.5782 0.6036 0.5819

Framework versions

  • Transformers 4.18.0
  • Pytorch 1.9.1
  • Datasets 2.1.0
  • Tokenizers 0.11.6