Boya1_RMSProp_1-e5_10Epoch_Swin-tiny-patch4-window16-256_fold4
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.1045
- Accuracy: 0.6318
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: 1e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
1.4543 | 1.0 | 924 | 1.4223 | 0.5088 |
1.2603 | 2.0 | 1848 | 1.2502 | 0.5733 |
1.0792 | 3.0 | 2772 | 1.1627 | 0.5939 |
1.0866 | 4.0 | 3696 | 1.1180 | 0.6183 |
0.9625 | 5.0 | 4620 | 1.0867 | 0.6212 |
0.6112 | 6.0 | 5544 | 1.0930 | 0.6234 |
0.774 | 7.0 | 6468 | 1.1000 | 0.6275 |
0.7257 | 8.0 | 7392 | 1.0961 | 0.6340 |
0.71 | 9.0 | 8316 | 1.1044 | 0.6291 |
0.5957 | 10.0 | 9240 | 1.1045 | 0.6318 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 14