metadata
library_name: transformers
license: apache-2.0
base_model: facebook/deit-base-patch16-224
tags:
- generated_from_trainer
metrics:
- accuracy
model-index:
- name: deit-base-patch16-224_rice-leaf-disease-augmented-v2_tl
results: []
deit-base-patch16-224_rice-leaf-disease-augmented-v2_tl
This model is a fine-tuned version of facebook/deit-base-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.8257
- Accuracy: 0.7232
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: 0.0003
- train_batch_size: 128
- eval_batch_size: 128
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 2.0445 | 1.0 | 63 | 1.8996 | 0.3631 |
| 1.6611 | 2.0 | 126 | 1.4965 | 0.5536 |
| 1.2741 | 3.0 | 189 | 1.2567 | 0.6190 |
| 1.0663 | 4.0 | 252 | 1.1184 | 0.6458 |
| 0.9481 | 5.0 | 315 | 1.0362 | 0.6696 |
| 0.8707 | 6.0 | 378 | 0.9921 | 0.6726 |
| 0.8161 | 7.0 | 441 | 0.9520 | 0.6875 |
| 0.7731 | 8.0 | 504 | 0.9180 | 0.6994 |
| 0.7417 | 9.0 | 567 | 0.8972 | 0.7054 |
| 0.7163 | 10.0 | 630 | 0.8765 | 0.7054 |
| 0.6954 | 11.0 | 693 | 0.8683 | 0.7054 |
| 0.6794 | 12.0 | 756 | 0.8527 | 0.7143 |
| 0.668 | 13.0 | 819 | 0.8441 | 0.7202 |
| 0.6575 | 14.0 | 882 | 0.8366 | 0.7202 |
| 0.65 | 15.0 | 945 | 0.8314 | 0.7173 |
| 0.645 | 16.0 | 1008 | 0.8284 | 0.7202 |
| 0.6417 | 17.0 | 1071 | 0.8270 | 0.7232 |
| 0.6402 | 18.0 | 1134 | 0.8262 | 0.7232 |
| 0.639 | 19.0 | 1197 | 0.8258 | 0.7232 |
| 0.6381 | 20.0 | 1260 | 0.8257 | 0.7232 |
Framework versions
- Transformers 4.48.3
- Pytorch 2.5.1+cu124
- Datasets 3.3.2
- Tokenizers 0.21.0