vit-base-patch16-224-in21k-mnist-100
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the mnist dataset. It achieves the following results on the evaluation set:
- Loss: 0.0290
- Accuracy: 0.9927
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: 2e-05
- train_batch_size: 32
- eval_batch_size: 256
- seed: 100
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.8271 | 1.0 | 1594 | 0.1402 | 0.9793 |
0.3895 | 2.0 | 3188 | 0.0672 | 0.9877 |
0.3556 | 3.0 | 4782 | 0.0515 | 0.9899 |
0.3272 | 4.0 | 6376 | 0.0401 | 0.9916 |
0.3084 | 5.0 | 7970 | 0.0388 | 0.9908 |
0.2978 | 6.0 | 9564 | 0.0336 | 0.992 |
0.2836 | 7.0 | 11158 | 0.0325 | 0.9922 |
0.2764 | 8.0 | 12752 | 0.0334 | 0.9918 |
0.258 | 9.0 | 14346 | 0.0297 | 0.992 |
0.2695 | 10.0 | 15940 | 0.0290 | 0.9927 |
Framework versions
- Transformers 4.28.0
- Pytorch 2.1.0
- Datasets 2.15.0
- Tokenizers 0.13.3
- Downloads last month
- 6