swin-finetuned-food101
This model is a fine-tuned version of microsoft/swin-base-patch4-window7-224 on the food101 dataset. It achieves the following results on the evaluation set:
- Loss: 0.2772
- Accuracy: 0.9210
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: 16
- eval_batch_size: 16
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 64
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.5077 | 1.0 | 1183 | 0.3851 | 0.8893 |
| 0.3523 | 2.0 | 2366 | 0.3124 | 0.9088 |
| 0.1158 | 3.0 | 3549 | 0.2772 | 0.9210 |
Framework versions
- Transformers 4.19.2
- Pytorch 1.11.0+cu113
- Datasets 2.2.2
- Tokenizers 0.12.1
- Downloads last month
- 62
Dataset used to train aspis/swin-finetuned-food101
Spaces using aspis/swin-finetuned-food101 2
Evaluation results
- Accuracy on food101self-reported0.921
- Accuracy on food101validation set self-reported0.914
- Precision Macro on food101validation set self-reported0.915
- Precision Micro on food101validation set self-reported0.914
- Precision Weighted on food101validation set self-reported0.915
- Recall Macro on food101validation set self-reported0.914
- Recall Micro on food101validation set self-reported0.914
- Recall Weighted on food101validation set self-reported0.914
- F1 Macro on food101validation set self-reported0.914
- F1 Micro on food101validation set self-reported0.914
- F1 Weighted on food101validation set self-reported0.914
- loss on food101validation set self-reported0.308