<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
distilbert-base-uncased-finetuned-emotion
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.2055
- Accuracy: 0.9355
- F1: 0.9354
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 | Accuracy | F1 |
---|---|---|---|---|---|
0.1775 | 1.0 | 250 | 0.1765 | 0.929 | 0.9287 |
0.1205 | 2.0 | 500 | 0.1516 | 0.9395 | 0.9393 |
0.0981 | 3.0 | 750 | 0.1530 | 0.9345 | 0.9351 |
0.0799 | 4.0 | 1000 | 0.1654 | 0.935 | 0.9348 |
0.0641 | 5.0 | 1250 | 0.1638 | 0.937 | 0.9364 |
0.0495 | 6.0 | 1500 | 0.1695 | 0.937 | 0.9369 |
0.0417 | 7.0 | 1750 | 0.1873 | 0.935 | 0.9350 |
0.0332 | 8.0 | 2000 | 0.1941 | 0.935 | 0.9351 |
0.0275 | 9.0 | 2250 | 0.1977 | 0.9385 | 0.9385 |
0.0224 | 10.0 | 2500 | 0.2055 | 0.9355 | 0.9354 |
Framework versions
- Transformers 4.13.0
- Pytorch 1.11.0
- Datasets 1.16.1
- Tokenizers 0.10.3