<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
vit-base-patch16-224-Trial006-YEL_STEM
This model is a fine-tuned version of google/vit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.1532
- Accuracy: 0.9655
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: 60
- eval_batch_size: 60
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 240
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 50
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6788 | 0.89 | 2 | 0.6730 | 0.6034 |
| 0.6825 | 1.78 | 4 | 0.6533 | 0.6724 |
| 0.6858 | 2.67 | 6 | 0.6266 | 0.6724 |
| 0.6423 | 4.0 | 9 | 0.5783 | 0.6897 |
| 0.6051 | 4.89 | 11 | 0.5173 | 0.7931 |
| 0.5636 | 5.78 | 13 | 0.4754 | 0.8103 |
| 0.5498 | 6.67 | 15 | 0.5737 | 0.7069 |
| 0.5432 | 8.0 | 18 | 0.4719 | 0.7759 |
| 0.5147 | 8.89 | 20 | 0.4669 | 0.7586 |
| 0.4892 | 9.78 | 22 | 0.4424 | 0.8103 |
| 0.486 | 10.67 | 24 | 0.4696 | 0.7586 |
| 0.4431 | 12.0 | 27 | 0.4678 | 0.7586 |
| 0.4537 | 12.89 | 29 | 0.4060 | 0.8103 |
| 0.4392 | 13.78 | 31 | 0.3795 | 0.7931 |
| 0.4559 | 14.67 | 33 | 0.3570 | 0.7759 |
| 0.413 | 16.0 | 36 | 0.3498 | 0.7931 |
| 0.4191 | 16.89 | 38 | 0.3393 | 0.8276 |
| 0.3822 | 17.78 | 40 | 0.3185 | 0.8621 |
| 0.2987 | 18.67 | 42 | 0.3019 | 0.8448 |
| 0.3749 | 20.0 | 45 | 0.2859 | 0.8966 |
| 0.4423 | 20.89 | 47 | 0.3855 | 0.7931 |
| 0.315 | 21.78 | 49 | 0.2763 | 0.8966 |
| 0.3141 | 22.67 | 51 | 0.2711 | 0.8793 |
| 0.3078 | 24.0 | 54 | 0.2180 | 0.9138 |
| 0.3052 | 24.89 | 56 | 0.2092 | 0.9138 |
| 0.319 | 25.78 | 58 | 0.2417 | 0.8793 |
| 0.3391 | 26.67 | 60 | 0.1940 | 0.9138 |
| 0.2742 | 28.0 | 63 | 0.1947 | 0.9138 |
| 0.332 | 28.89 | 65 | 0.1938 | 0.8793 |
| 0.3181 | 29.78 | 67 | 0.1629 | 0.9310 |
| 0.3294 | 30.67 | 69 | 0.1623 | 0.9138 |
| 0.2832 | 32.0 | 72 | 0.2068 | 0.9138 |
| 0.3214 | 32.89 | 74 | 0.1799 | 0.9138 |
| 0.2682 | 33.78 | 76 | 0.1645 | 0.9138 |
| 0.288 | 34.67 | 78 | 0.1627 | 0.9138 |
| 0.3077 | 36.0 | 81 | 0.1616 | 0.9310 |
| 0.2459 | 36.89 | 83 | 0.1504 | 0.9138 |
| 0.2405 | 37.78 | 85 | 0.1532 | 0.9655 |
| 0.2911 | 38.67 | 87 | 0.1548 | 0.9310 |
| 0.2545 | 40.0 | 90 | 0.1573 | 0.9138 |
| 0.25 | 40.89 | 92 | 0.1574 | 0.9310 |
| 0.2633 | 41.78 | 94 | 0.1581 | 0.9310 |
| 0.2765 | 42.67 | 96 | 0.1627 | 0.9138 |
| 0.263 | 44.0 | 99 | 0.1605 | 0.9138 |
| 0.2904 | 44.44 | 100 | 0.1597 | 0.9138 |
Framework versions
- Transformers 4.30.0.dev0
- Pytorch 1.12.1
- Datasets 2.12.0
- Tokenizers 0.13.1
