vit-base-patch16-224-in21k-face-recognition
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.0015
- Accuracy: 1.0000
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: 0.00012
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 256
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.0368 | 1.0 | 372 | 0.0346 | 1.0000 |
| 0.0094 | 2.0 | 744 | 0.0092 | 1.0000 |
| 0.0046 | 3.0 | 1116 | 0.0047 | 1.0000 |
| 0.0029 | 4.0 | 1488 | 0.0029 | 1.0 |
| 0.0022 | 5.0 | 1860 | 0.0023 | 0.9999 |
| 0.0017 | 6.0 | 2232 | 0.0017 | 1.0 |
| 0.0015 | 7.0 | 2604 | 0.0015 | 1.0 |
| 0.0014 | 8.0 | 2976 | 0.0015 | 1.0000 |
Framework versions
- Transformers 4.30.2
- Pytorch 1.13.1+cu117
- Datasets 2.13.2
- Tokenizers 0.11.0