metadata
library_name: transformers
license: apache-2.0
base_model: google/vit-base-patch16-224-in21k
tags:
- image-classification
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: finetuned-fake-food
results: []
finetuned-fake-food
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the indian_food_images dataset. It achieves the following results on the evaluation set:
- Loss: 0.6574
- Accuracy: 0.6164
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: 2
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6977 | 0.2421 | 100 | 0.6805 | 0.5822 |
| 0.6956 | 0.4843 | 200 | 0.6936 | 0.4178 |
| 0.6795 | 0.7264 | 300 | 0.6734 | 0.6507 |
| 0.7061 | 0.9685 | 400 | 0.6760 | 0.5822 |
| 0.6941 | 1.2107 | 500 | 0.6746 | 0.5822 |
| 0.6898 | 1.4528 | 600 | 0.6675 | 0.6027 |
| 0.6956 | 1.6949 | 700 | 0.6846 | 0.5753 |
| 0.6847 | 1.9370 | 800 | 0.6746 | 0.5822 |
| 0.6949 | 2.1792 | 900 | 0.6780 | 0.5890 |
| 0.703 | 2.4213 | 1000 | 0.6895 | 0.5753 |
| 0.6851 | 2.6634 | 1100 | 0.6742 | 0.5822 |
| 0.6878 | 2.9056 | 1200 | 0.6742 | 0.6301 |
| 0.68 | 3.1477 | 1300 | 0.6713 | 0.5822 |
| 0.6728 | 3.3898 | 1400 | 0.6838 | 0.5959 |
| 0.698 | 3.6320 | 1500 | 0.6775 | 0.5822 |
| 0.7033 | 3.8741 | 1600 | 0.6735 | 0.5822 |
| 0.6973 | 4.1162 | 1700 | 0.6804 | 0.6233 |
| 0.6822 | 4.3584 | 1800 | 0.6848 | 0.6027 |
| 0.6896 | 4.6005 | 1900 | 0.6835 | 0.5411 |
| 0.6772 | 4.8426 | 2000 | 0.6753 | 0.6096 |
| 0.6843 | 5.0847 | 2100 | 0.6667 | 0.5890 |
| 0.6898 | 5.3269 | 2200 | 0.6726 | 0.5822 |
| 0.6868 | 5.5690 | 2300 | 0.6784 | 0.5616 |
| 0.6636 | 5.8111 | 2400 | 0.6640 | 0.6301 |
| 0.6833 | 6.0533 | 2500 | 0.6768 | 0.5137 |
| 0.678 | 6.2954 | 2600 | 0.6652 | 0.6233 |
| 0.6672 | 6.5375 | 2700 | 0.6735 | 0.5479 |
| 0.6975 | 6.7797 | 2800 | 0.6687 | 0.5890 |
| 0.6858 | 7.0218 | 2900 | 0.6672 | 0.6027 |
| 0.6687 | 7.2639 | 3000 | 0.6648 | 0.5753 |
| 0.6636 | 7.5061 | 3100 | 0.6674 | 0.5685 |
| 0.6904 | 7.7482 | 3200 | 0.6752 | 0.5342 |
| 0.6585 | 7.9903 | 3300 | 0.7023 | 0.5959 |
| 0.6874 | 8.2324 | 3400 | 0.6615 | 0.5753 |
| 0.6444 | 8.4746 | 3500 | 0.7721 | 0.5205 |
| 0.6803 | 8.7167 | 3600 | 0.6809 | 0.5822 |
| 0.6782 | 8.9588 | 3700 | 0.6638 | 0.5822 |
| 0.6536 | 9.2010 | 3800 | 0.6607 | 0.6233 |
| 0.6188 | 9.4431 | 3900 | 0.7090 | 0.5685 |
| 0.7026 | 9.6852 | 4000 | 0.6574 | 0.6164 |
| 0.7008 | 9.9274 | 4100 | 0.6577 | 0.6096 |
Framework versions
- Transformers 4.44.2
- Pytorch 2.4.1+cu121
- Datasets 3.0.1
- Tokenizers 0.19.1