jaffe_V2_20_1
This model is a fine-tuned version of WinKawaks/vit-tiny-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.5014
- Accuracy: 0.4333
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 1 | 2.4111 | 0.1333 |
| No log | 2.0 | 2 | 2.0318 | 0.1667 |
| No log | 3.0 | 3 | 2.0433 | 0.1667 |
| No log | 4.0 | 4 | 2.0298 | 0.1 |
| No log | 5.0 | 5 | 1.9400 | 0.2 |
| No log | 6.0 | 6 | 1.8599 | 0.2667 |
| No log | 7.0 | 7 | 1.7569 | 0.2333 |
| No log | 8.0 | 8 | 1.8134 | 0.1333 |
| No log | 9.0 | 9 | 1.6201 | 0.3667 |
| 1.971 | 10.0 | 10 | 1.6919 | 0.2667 |
| 1.971 | 11.0 | 11 | 1.6849 | 0.2667 |
| 1.971 | 12.0 | 12 | 1.6344 | 0.3333 |
| 1.971 | 13.0 | 13 | 1.6805 | 0.3 |
| 1.971 | 14.0 | 14 | 1.5784 | 0.4667 |
| 1.971 | 15.0 | 15 | 1.5104 | 0.4667 |
| 1.971 | 16.0 | 16 | 1.4978 | 0.4333 |
| 1.971 | 17.0 | 17 | 1.5169 | 0.4 |
| 1.971 | 18.0 | 18 | 1.4790 | 0.4 |
| 1.971 | 19.0 | 19 | 1.4223 | 0.6 |
| 1.5094 | 20.0 | 20 | 1.5014 | 0.4333 |
Framework versions
- Transformers 4.47.0
- Pytorch 2.5.1+cu121
- Datasets 3.3.1
- Tokenizers 0.21.0