vit-base-patch16-224-in21k-food101-24-12
This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the food101 dataset. It achieves the following results on the evaluation set:
- Loss: 0.3533
- Accuracy: 0.9069
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: 24
- eval_batch_size: 24
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 96
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 12
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy |
---|---|---|---|---|
2.7927 | 1.0 | 789 | 2.5629 | 0.7693 |
1.256 | 2.0 | 1578 | 0.9637 | 0.8583 |
0.94 | 3.0 | 2367 | 0.5866 | 0.8816 |
0.6693 | 4.0 | 3157 | 0.4752 | 0.8888 |
0.6337 | 5.0 | 3946 | 0.4282 | 0.8941 |
0.5811 | 6.0 | 4735 | 0.4110 | 0.8949 |
0.4661 | 7.0 | 5524 | 0.3875 | 0.8990 |
0.4188 | 8.0 | 6314 | 0.3776 | 0.9010 |
0.5045 | 9.0 | 7103 | 0.3633 | 0.9049 |
0.3437 | 10.0 | 7892 | 0.3611 | 0.9058 |
0.3494 | 11.0 | 8681 | 0.3568 | 0.9060 |
0.3381 | 12.0 | 9468 | 0.3533 | 0.9069 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1
- Downloads last month
- 6
Inference Providers
NEW
This model is not currently available via any of the supported third-party Inference Providers, and
the model is not deployed on the HF Inference API.
Model tree for PedroSampaio/vit-base-patch16-224-in21k-food101-24-12
Base model
google/vit-base-patch16-224-in21k