fer_plus_V2
This model is a fine-tuned version of microsoft/beit-base-patch16-224-pt22k-ft22k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.7483
- Accuracy: 0.7598
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0697 | 1.0 | 222 | 1.0167 | 0.6354 |
| 0.7784 | 2.0 | 444 | 0.8059 | 0.7124 |
| 0.5911 | 3.0 | 666 | 0.7499 | 0.7384 |
| 0.4609 | 4.0 | 888 | 0.7586 | 0.7502 |
| 0.3712 | 5.0 | 1110 | 0.7483 | 0.7598 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.3.1
- Tokenizers 0.21.0