Boya1_RMSProp_1-e5_10Epoch_swin-large-patch4-window7-224_fold2
This model is a fine-tuned version of microsoft/swin-large-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 1.1478
- Accuracy: 0.6735
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 |
---|---|---|---|---|
1.2127 | 1.0 | 923 | 1.1541 | 0.6049 |
0.9204 | 2.0 | 1846 | 1.0192 | 0.6551 |
0.7902 | 3.0 | 2769 | 0.9565 | 0.6649 |
0.7847 | 4.0 | 3692 | 0.9757 | 0.6743 |
0.6816 | 5.0 | 4615 | 0.9792 | 0.6692 |
0.4595 | 6.0 | 5538 | 1.0390 | 0.6678 |
0.425 | 7.0 | 6461 | 1.0878 | 0.6678 |
0.4186 | 8.0 | 7384 | 1.1180 | 0.6727 |
0.3048 | 9.0 | 8307 | 1.1393 | 0.6708 |
0.2135 | 10.0 | 9230 | 1.1478 | 0.6735 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 6