deit-base-patch16-224-finetuned-ind-4-imbalanced-aadhaarmask-3839
This model is a fine-tuned version of facebook/deit-base-patch16-224 on the imagefolder dataset. It achieves the following results on the evaluation set:
- Loss: 0.2241
- Accuracy: 0.9401
- Recall: 0.9401
- F1: 0.9385
- Precision: 0.9382
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: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- 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 | Recall | F1 | Precision |
|---|---|---|---|---|---|---|---|
| 0.1426 | 1.0 | 96 | 0.2195 | 0.9297 | 0.9297 | 0.9263 | 0.9270 |
| 0.0644 | 2.0 | 192 | 0.2403 | 0.9245 | 0.9245 | 0.9249 | 0.9260 |
| 0.0695 | 3.0 | 288 | 0.3488 | 0.9232 | 0.9232 | 0.9221 | 0.9257 |
| 0.0674 | 4.0 | 384 | 0.2355 | 0.9375 | 0.9375 | 0.9366 | 0.9363 |
| 0.1265 | 5.0 | 480 | 0.2119 | 0.9388 | 0.9388 | 0.9376 | 0.9382 |
| 0.1128 | 6.0 | 576 | 0.2018 | 0.9401 | 0.9401 | 0.9388 | 0.9389 |
| 0.0806 | 7.0 | 672 | 0.2095 | 0.9388 | 0.9388 | 0.9371 | 0.9410 |
| 0.1237 | 8.0 | 768 | 0.2008 | 0.9427 | 0.9427 | 0.9423 | 0.9425 |
| 0.0955 | 9.0 | 864 | 0.1763 | 0.9440 | 0.9440 | 0.9420 | 0.9421 |
| 0.0429 | 10.0 | 960 | 0.2021 | 0.9401 | 0.9401 | 0.9381 | 0.9376 |
Framework versions
- Transformers 4.40.1
- Pytorch 2.2.1+cu121
- Datasets 2.19.0
- Tokenizers 0.19.1