swinv2-tiny-patch4-window8-256-finetuned-ind-17-imbalanced-aadhaarmask
This model is a fine-tuned version of microsoft/swinv2-tiny-patch4-window8-256 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.3601
- Accuracy: 0.8565
- Recall: 0.8565
- F1: 0.8537
- Precision: 0.8631
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 |
|---|---|---|---|---|---|---|---|
| No log | 0.9974 | 293 | 0.6645 | 0.7820 | 0.7820 | 0.7661 | 0.7678 |
| No log | 1.9983 | 587 | 0.5493 | 0.8033 | 0.8033 | 0.7897 | 0.7964 |
| No log | 2.9991 | 881 | 0.4242 | 0.8416 | 0.8416 | 0.8380 | 0.8460 |
| No log | 4.0 | 1175 | 0.4124 | 0.8310 | 0.8310 | 0.8288 | 0.8299 |
| No log | 4.9974 | 1468 | 0.3769 | 0.8412 | 0.8412 | 0.8388 | 0.8478 |
| No log | 5.9983 | 1762 | 0.3589 | 0.8501 | 0.8501 | 0.8481 | 0.8582 |
| No log | 6.9991 | 2056 | 0.3503 | 0.8455 | 0.8455 | 0.8456 | 0.8535 |
| No log | 8.0 | 2350 | 0.3400 | 0.8404 | 0.8404 | 0.8416 | 0.8465 |
| No log | 8.9974 | 2643 | 0.3533 | 0.8480 | 0.8480 | 0.8480 | 0.8501 |
| 0.5214 | 9.9745 | 2930 | 0.3358 | 0.8459 | 0.8459 | 0.8460 | 0.8473 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.0a0+81ea7a4
- Datasets 2.19.0
- Tokenizers 0.19.1