swin-base-patch4-window7-224-finetuned-ind-17-imbalanced-aadhaarmask
This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3209
- Accuracy: 0.8557
- Recall: 0.8557
- F1: 0.8542
- Precision: 0.8560
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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 | Recall | F1 | Precision |
|---|---|---|---|---|---|---|---|
| 0.5155 | 0.9974 | 293 | 0.5710 | 0.7935 | 0.7935 | 0.7821 | 0.7895 |
| 0.4245 | 1.9983 | 587 | 0.4729 | 0.8238 | 0.8238 | 0.8187 | 0.8266 |
| 0.4183 | 2.9991 | 881 | 0.4145 | 0.8408 | 0.8408 | 0.8309 | 0.8350 |
| 0.4088 | 4.0 | 1175 | 0.3901 | 0.8425 | 0.8425 | 0.8375 | 0.8501 |
| 0.3489 | 4.9974 | 1468 | 0.3703 | 0.8463 | 0.8463 | 0.8446 | 0.8518 |
| 0.3115 | 5.9983 | 1762 | 0.3500 | 0.8540 | 0.8540 | 0.8525 | 0.8605 |
| 0.3087 | 6.9991 | 2056 | 0.3338 | 0.8519 | 0.8519 | 0.8494 | 0.8582 |
| 0.2372 | 8.0 | 2350 | 0.3181 | 0.8548 | 0.8548 | 0.8543 | 0.8587 |
| 0.2816 | 8.9974 | 2643 | 0.3167 | 0.8536 | 0.8536 | 0.8530 | 0.8561 |
| 0.2378 | 9.9745 | 2930 | 0.3063 | 0.8702 | 0.8702 | 0.8686 | 0.8709 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.0a0+81ea7a4
- Datasets 2.19.0
- Tokenizers 0.19.1