swin-tiny-patch4-window7-224-finetuned-fluro_cls
This model is a fine-tuned version of microsoft/swin-tiny-patch4-window7-224 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.0000
- Accuracy: 1.0
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 | 0.67 | 1 | 0.7112 | 0.5238 |
| No log | 1.67 | 2 | 0.5591 | 0.8571 |
| 0.811 | 2.67 | 3 | 0.3781 | 0.9524 |
| 0.811 | 3.67 | 4 | 0.1995 | 1.0 |
| 0.811 | 4.67 | 5 | 0.1215 | 1.0 |
| 0.3531 | 5.67 | 6 | 0.0578 | 1.0 |
| 0.3531 | 6.67 | 7 | 0.0195 | 1.0 |
| 0.3531 | 7.67 | 8 | 0.0072 | 1.0 |
| 0.0618 | 8.67 | 9 | 0.0030 | 1.0 |
| 0.0618 | 9.67 | 10 | 0.0012 | 1.0 |
| 0.0618 | 10.67 | 11 | 0.0005 | 1.0 |
| 0.0079 | 11.67 | 12 | 0.0003 | 1.0 |
| 0.0079 | 12.67 | 13 | 0.0001 | 1.0 |
| 0.0079 | 13.67 | 14 | 0.0001 | 1.0 |
| 0.0051 | 14.67 | 15 | 0.0001 | 1.0 |
| 0.0051 | 15.67 | 16 | 0.0000 | 1.0 |
| 0.0051 | 16.67 | 17 | 0.0000 | 1.0 |
| 0.0017 | 17.67 | 18 | 0.0000 | 1.0 |
| 0.0017 | 18.67 | 19 | 0.0000 | 1.0 |
| 0.0017 | 19.67 | 20 | 0.0000 | 1.0 |
| 0.0004 | 20.67 | 21 | 0.0000 | 1.0 |
| 0.0004 | 21.67 | 22 | 0.0000 | 1.0 |
| 0.0004 | 22.67 | 23 | 0.0000 | 1.0 |
| 0.0022 | 23.67 | 24 | 0.0000 | 1.0 |
| 0.0022 | 24.67 | 25 | 0.0000 | 1.0 |
| 0.0022 | 25.67 | 26 | 0.0000 | 1.0 |
| 0.001 | 26.67 | 27 | 0.0000 | 1.0 |
| 0.001 | 27.67 | 28 | 0.0000 | 1.0 |
| 0.001 | 28.67 | 29 | 0.0000 | 1.0 |
| 0.0013 | 29.67 | 30 | 0.0000 | 1.0 |
Framework versions
- Transformers 4.25.1
- Pytorch 1.10.2+cu113
- Datasets 2.7.1
- Tokenizers 0.13.2