Boya1_RMSProp_1-e5_10Epoch_Swin-tiny-patch4-window16-256_fold5
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: 1.1116
- Accuracy: 0.6365
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 |
---|---|---|---|---|
1.5163 | 1.0 | 924 | 1.5369 | 0.4812 |
1.3319 | 2.0 | 1848 | 1.2545 | 0.5747 |
1.0647 | 3.0 | 2772 | 1.1479 | 0.6031 |
0.8661 | 4.0 | 3696 | 1.1242 | 0.6183 |
1.0965 | 5.0 | 4620 | 1.1043 | 0.6235 |
0.9375 | 6.0 | 5544 | 1.1024 | 0.6311 |
0.7507 | 7.0 | 6468 | 1.1093 | 0.6313 |
0.6725 | 8.0 | 7392 | 1.1187 | 0.6286 |
0.5695 | 9.0 | 8316 | 1.1151 | 0.6354 |
0.6509 | 10.0 | 9240 | 1.1116 | 0.6365 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 14