metadata
license: apache-2.0
tags:
- generated_from_trainer
datasets:
- emotion
model-index:
- name: distilbert-base-uncased-fine-tuned-emotions
results: []
distilbert-base-uncased-fine-tuned-emotions
This model is a fine-tuned version of distilbert-base-uncased on the emotion dataset. It achieves the following results on the evaluation set:
- Loss: 0.2433
- Acc : 0.9355
- F1 : 0.9356
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: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Acc | F1 |
|---|---|---|---|---|---|
| 0.1674 | 1.0 | 250 | 0.1948 | 0.9275 | 0.9267 |
| 0.1185 | 2.0 | 500 | 0.1635 | 0.938 | 0.9380 |
| 0.0998 | 3.0 | 750 | 0.1723 | 0.9345 | 0.9352 |
| 0.0808 | 4.0 | 1000 | 0.1687 | 0.934 | 0.9337 |
| 0.0621 | 5.0 | 1250 | 0.1769 | 0.937 | 0.9368 |
| 0.0511 | 6.0 | 1500 | 0.1927 | 0.933 | 0.9327 |
| 0.0393 | 7.0 | 1750 | 0.2275 | 0.9345 | 0.9348 |
| 0.0323 | 8.0 | 2000 | 0.2338 | 0.932 | 0.9324 |
| 0.0273 | 9.0 | 2250 | 0.2439 | 0.935 | 0.9352 |
| 0.0215 | 10.0 | 2500 | 0.2433 | 0.9355 | 0.9356 |
Framework versions
- Transformers 4.27.2
- Pytorch 1.13.1+cu117
- Datasets 2.11.0
- Tokenizers 0.13.3