Hugging Face's logo

philschmid
/
minilm-l12-h384-sst2-distilled

Text Classification
Transformers
PyTorch
TensorBoard
roberta
Generated from Trainer
Eval Results
Inference Endpoints
Model card Files Metrics Community
3
minilm-l12-h384-sst2-distilled / README.md
philschmid's picture
philschmid HF staff
End of training
82ceba1
preview code
|
raw
history blame contribute delete
1.95 kB
metadata
tags:
  - generated_from_trainer
datasets:
  - glue
metrics:
  - accuracy
model-index:
  - name: minilm-l12-h384-sst2-distilled
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: glue
          type: glue
          args: sst2
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9220183486238532

minilm-l12-h384-sst2-distilled

This model is a fine-tuned version of nreimers/MiniLMv2-L12-H384-distilled-from-RoBERTa-Large on the glue dataset. It achieves the following results on the evaluation set:

  • Loss: 0.5417
  • Accuracy: 0.9220

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: 0.0001400785945474408
  • train_batch_size: 512
  • eval_batch_size: 512
  • seed: 33
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • num_epochs: 5
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
2.2689 1.0 132 0.7102 0.8979
0.8295 2.0 264 0.5669 0.9117
0.5059 3.0 396 0.5545 0.9220
0.3722 4.0 528 0.5378 0.9209
0.2924 5.0 660 0.5417 0.9220

Framework versions

  • Transformers 4.17.0
  • Pytorch 1.10.2+cu113
  • Datasets 1.18.4
  • Tokenizers 0.11.6