swinv2-tiny-patch4-window8-256-finetuned-gardner-te-max
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: 0.8795
- Accuracy: 0.5940
Model description
Predict Trophectoderm Grade - Gardner Score from an embryo image
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: 20
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.0943 | 0.94 | 11 | 1.0750 | 0.6325 |
| 0.9996 | 1.96 | 23 | 0.8011 | 0.6325 |
| 0.7731 | 2.98 | 35 | 0.7182 | 0.6325 |
| 0.7564 | 4.0 | 47 | 0.7109 | 0.6325 |
| 0.7331 | 4.94 | 58 | 0.7026 | 0.6325 |
| 0.7336 | 5.96 | 70 | 0.6848 | 0.6325 |
| 0.7305 | 6.98 | 82 | 0.6938 | 0.6325 |
| 0.7314 | 8.0 | 94 | 0.6549 | 0.6325 |
| 0.6905 | 8.94 | 105 | 0.6364 | 0.6867 |
| 0.7315 | 9.96 | 117 | 0.6223 | 0.6687 |
| 0.6839 | 10.98 | 129 | 0.6528 | 0.7530 |
| 0.6931 | 12.0 | 141 | 0.6209 | 0.7410 |
| 0.6705 | 12.94 | 152 | 0.6296 | 0.7169 |
| 0.7227 | 13.96 | 164 | 0.6039 | 0.7108 |
| 0.6695 | 14.98 | 176 | 0.6049 | 0.7530 |
| 0.6981 | 16.0 | 188 | 0.5965 | 0.7048 |
| 0.6566 | 16.94 | 199 | 0.6111 | 0.7410 |
| 0.6828 | 17.96 | 211 | 0.5969 | 0.7530 |
| 0.6632 | 18.72 | 220 | 0.5947 | 0.7530 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.2
- Datasets 2.16.0
- Tokenizers 0.15.0