swin-tiny-patch4-window7-224-finetuned-eurosat-leukemia-3000
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: 0.3623
- Accuracy: 0.865
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: 10
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
0.5519 | 0.9941 | 42 | 0.7743 | 0.5983 |
0.4142 | 1.9882 | 84 | 0.4881 | 0.7267 |
0.3891 | 2.9822 | 126 | 0.3352 | 0.8483 |
0.3091 | 4.0 | 169 | 0.3623 | 0.865 |
0.296 | 4.9941 | 211 | 0.7635 | 0.7183 |
0.2561 | 5.9882 | 253 | 0.5991 | 0.7717 |
0.2292 | 6.9822 | 295 | 0.6934 | 0.745 |
0.2039 | 8.0 | 338 | 0.7756 | 0.7267 |
0.1969 | 8.9941 | 380 | 0.5697 | 0.7867 |
0.1845 | 9.9408 | 420 | 0.5544 | 0.79 |
Framework versions
- Transformers 4.40.0.dev0
- Pytorch 2.2.2+cpu
- Datasets 2.19.0
- Tokenizers 0.15.2
- Downloads last month
- 16