vit-dropout-v6
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.4214
- Accuracy: 0.8633
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: 0.0001
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 8
- mixed_precision_training: Native AMP
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.4882 | 1.56 | 500 | 0.8790 | 0.7247 |
0.3687 | 3.12 | 1000 | 0.5066 | 0.8140 |
0.3441 | 4.67 | 1500 | 0.4167 | 0.8502 |
0.3001 | 6.23 | 2000 | 0.4363 | 0.8583 |
0.2672 | 7.79 | 2500 | 0.4214 | 0.8633 |
Framework versions
- Transformers 4.38.1
- Pytorch 2.1.0+cu121
- Datasets 2.17.1
- Tokenizers 0.15.2
- Downloads last month
- 3