Hugging Face's logo

philschmid
/
MiniLMv2-L6-H384-emotion

Text Classification Transformers PyTorch
roberta generated_from_trainer Eval Results Inference Endpoints
Model card Files Community
1
MiniLMv2-L6-H384-emotion / README.md
philschmid's picture
philschmid HF staff
End of training
96350ab
preview code
|
raw history blame contribute delete
No virus
2.14 kB
metadata
tags:
  - generated_from_trainer
datasets:
  - emotion
metrics:
  - accuracy
model-index:
  - name: MiniLMv2-L6-H384-emotion
    results:
      - task:
          name: Text Classification
          type: text-classification
        dataset:
          name: emotion
          type: emotion
          args: default
        metrics:
          - name: Accuracy
            type: accuracy
            value: 0.9215

MiniLMv2-L6-H384-emotion

This model is a fine-tuned version of nreimers/MiniLMv2-L6-H384-distilled-from-RoBERTa-Large on the emotion dataset. It achieves the following results on the evaluation set:

  • Loss: 0.2140
  • Accuracy: 0.9215

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: 3e-05
  • train_batch_size: 32
  • eval_batch_size: 32
  • seed: 42
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_steps: 500
  • num_epochs: 8
  • mixed_precision_training: Native AMP

Training results

Training Loss Epoch Step Validation Loss Accuracy
1.432 1.0 500 0.9992 0.6805
0.8073 2.0 1000 0.5437 0.846
0.4483 3.0 1500 0.3018 0.909
0.2833 4.0 2000 0.2412 0.915
0.2169 5.0 2500 0.2140 0.9215
0.1821 6.0 3000 0.2159 0.917
0.154 7.0 3500 0.2084 0.919
0.1461 8.0 4000 0.2047 0.92

Framework versions

  • Transformers 4.12.3
  • Pytorch 1.9.1
  • Datasets 1.15.1
  • Tokenizers 0.10.3