compoundleaf-swin-tiny-patch4-window7-224-finetuned
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: 0.0204
- Accuracy: 0.9955
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 7 | 0.9808 | 0.5446 |
| 1.1008 | 2.0 | 14 | 0.6537 | 0.8438 |
| 0.6618 | 3.0 | 21 | 0.2581 | 0.9464 |
| 0.6618 | 4.0 | 28 | 0.0740 | 0.9866 |
| 0.1685 | 5.0 | 35 | 0.0405 | 0.9911 |
| 0.0552 | 6.0 | 42 | 0.0824 | 0.9866 |
| 0.0552 | 7.0 | 49 | 0.0753 | 0.9777 |
| 0.0422 | 8.0 | 56 | 0.0204 | 0.9955 |
| 0.0212 | 9.0 | 63 | 0.0646 | 0.9821 |
| 0.02 | 10.0 | 70 | 0.0172 | 0.9955 |
| 0.02 | 11.0 | 77 | 0.0279 | 0.9955 |
| 0.0159 | 12.0 | 84 | 0.0221 | 0.9955 |
| 0.0053 | 13.0 | 91 | 0.0415 | 0.9866 |
| 0.0053 | 14.0 | 98 | 0.0222 | 0.9955 |
| 0.0075 | 15.0 | 105 | 0.0408 | 0.9955 |
| 0.003 | 16.0 | 112 | 0.0187 | 0.9955 |
| 0.003 | 17.0 | 119 | 0.0307 | 0.9911 |
| 0.0073 | 18.0 | 126 | 0.0327 | 0.9955 |
| 0.0079 | 19.0 | 133 | 0.0203 | 0.9955 |
| 0.0066 | 20.0 | 140 | 0.0319 | 0.9955 |
| 0.0066 | 21.0 | 147 | 0.0451 | 0.9866 |
| 0.0038 | 22.0 | 154 | 0.0264 | 0.9955 |
| 0.0122 | 23.0 | 161 | 0.0968 | 0.9821 |
| 0.0122 | 24.0 | 168 | 0.0547 | 0.9911 |
| 0.0096 | 25.0 | 175 | 0.0260 | 0.9955 |
| 0.0034 | 26.0 | 182 | 0.0432 | 0.9911 |
| 0.0034 | 27.0 | 189 | 0.0587 | 0.9911 |
| 0.009 | 28.0 | 196 | 0.0478 | 0.9911 |
| 0.004 | 29.0 | 203 | 0.0323 | 0.9955 |
| 0.0081 | 30.0 | 210 | 0.0302 | 0.9955 |
Framework versions
- Transformers 4.38.1
- Pytorch 1.10.0+cu111
- Datasets 2.17.1
- Tokenizers 0.15.2