hushem_40x_deit_small_f1
This model is a fine-tuned version of facebook/deit-small-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.7923
- Accuracy: 0.8
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: 32
- eval_batch_size: 32
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.1638 | 0.99 | 53 | 0.4948 | 0.8222 |
| 0.018 | 1.99 | 107 | 0.8208 | 0.7556 |
| 0.0086 | 3.0 | 161 | 0.6473 | 0.8667 |
| 0.0011 | 4.0 | 215 | 0.7960 | 0.7556 |
| 0.0003 | 4.99 | 268 | 0.8013 | 0.7556 |
| 0.0001 | 5.99 | 322 | 0.8035 | 0.8 |
| 0.0001 | 7.0 | 376 | 0.7952 | 0.8 |
| 0.0001 | 8.0 | 430 | 0.7939 | 0.8 |
| 0.0001 | 8.99 | 483 | 0.7931 | 0.8 |
| 0.0001 | 9.86 | 530 | 0.7923 | 0.8 |
Framework versions
- Transformers 4.35.0
- Pytorch 2.1.0+cu118
- Datasets 2.14.6
- Tokenizers 0.14.1