distilbert-base-uncased-finetuned-emotion
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: 0.1443
- Accuracy: 0.9375
- F1: 0.9378
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: 256
- eval_batch_size: 256
- 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 | Accuracy | F1 |
---|---|---|---|---|---|
No log | 1.0 | 63 | 0.8657 | 0.714 | 0.6520 |
No log | 2.0 | 126 | 0.3186 | 0.9085 | 0.9076 |
No log | 3.0 | 189 | 0.2032 | 0.928 | 0.9281 |
0.5856 | 4.0 | 252 | 0.1733 | 0.93 | 0.9301 |
0.5856 | 5.0 | 315 | 0.1578 | 0.937 | 0.9368 |
0.5856 | 6.0 | 378 | 0.1543 | 0.9335 | 0.9341 |
0.5856 | 7.0 | 441 | 0.1506 | 0.9345 | 0.9343 |
0.1139 | 8.0 | 504 | 0.1475 | 0.939 | 0.9396 |
0.1139 | 9.0 | 567 | 0.1444 | 0.9375 | 0.9374 |
0.1139 | 10.0 | 630 | 0.1443 | 0.9375 | 0.9378 |
Framework versions
- Transformers 4.41.2
- Pytorch 2.3.1+cu121
- Datasets 2.20.0
- Tokenizers 0.19.1
- Downloads last month
- 5
Finetuned from
Dataset used to train yspkm/distilbert-base-uncased-finetuned-emotion
Evaluation results
- Accuracy on emotionvalidation set self-reported0.938
- F1 on emotionvalidation set self-reported0.938