vit-base-age-classification
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the fair_face dataset. It achieves the following results on the evaluation set:
- Loss: 0.0743
- Accuracy: 0.9879
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: 0.0002
- train_batch_size: 32
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.2011 | 1.0 | 385 | 1.0297 | 0.5664 |
| 0.8578 | 2.0 | 770 | 0.7667 | 0.6936 |
| 0.5961 | 3.0 | 1155 | 0.4088 | 0.8703 |
| 0.3073 | 4.0 | 1540 | 0.1689 | 0.9581 |
| 0.1146 | 5.0 | 1925 | 0.0743 | 0.9879 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1