vit-base-breast-cancer
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the breast-invasive-ductal-carcinoma-cancer dataset. It achieves the following results on the evaluation set:
- Loss: 0.3578
- Accuracy: 0.8731
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: 0.0002
- train_batch_size: 16
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 4
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.206 | 0.23 | 100 | 0.3991 | 0.8342 |
| 0.2959 | 0.46 | 200 | 0.3608 | 0.8691 |
| 0.2492 | 0.69 | 300 | 0.3624 | 0.8681 |
| 0.237 | 0.92 | 400 | 0.3578 | 0.8731 |
| 0.2277 | 1.14 | 500 | 0.4044 | 0.8721 |
| 0.1821 | 1.37 | 600 | 0.4185 | 0.8661 |
| 0.1995 | 1.6 | 700 | 0.4412 | 0.8721 |
| 0.1288 | 1.83 | 800 | 0.4424 | 0.8272 |
| 0.1546 | 2.06 | 900 | 0.4491 | 0.8641 |
| 0.0319 | 2.29 | 1000 | 0.5021 | 0.8641 |
| 0.0537 | 2.52 | 1100 | 0.5292 | 0.8472 |
| 0.1493 | 2.75 | 1200 | 0.4235 | 0.8821 |
| 0.0774 | 2.97 | 1300 | 0.4753 | 0.8631 |
| 0.0823 | 3.2 | 1400 | 0.5120 | 0.8561 |
| 0.0264 | 3.43 | 1500 | 0.4868 | 0.8701 |
| 0.0386 | 3.66 | 1600 | 0.4909 | 0.8651 |
| 0.008 | 3.89 | 1700 | 0.5035 | 0.8651 |
Framework versions
- Transformers 4.27.1
- Pytorch 2.0.1+cu118
- Datasets 2.9.0
- Tokenizers 0.13.3