metadata
license: apache-2.0
base_model: google/vit-base-patch16-224
tags:
- generated_from_trainer
metrics:
- accuracy
- f1
- recall
- precision
model-index:
- name: vit-real-fake-classification-v4
results: []
vit-real-fake-classification-v4
This model is a fine-tuned version of google/vit-base-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.0585
- Accuracy: 0.9796
- F1: 0.9815
- Recall: 0.9815
- Precision: 0.9815
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: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Recall | Precision |
|---|---|---|---|---|---|---|---|
| 0.1295 | 1.0 | 233 | 0.2414 | 0.9151 | 0.9280 | 0.9912 | 0.8723 |
| 0.4466 | 2.0 | 466 | 0.1042 | 0.9646 | 0.9680 | 0.9718 | 0.9643 |
| 0.3302 | 3.0 | 699 | 0.0667 | 0.9764 | 0.9786 | 0.9776 | 0.9795 |
| 0.0003 | 4.0 | 932 | 0.0995 | 0.9731 | 0.9758 | 0.9796 | 0.9720 |
| 0.0002 | 5.0 | 1165 | 0.0585 | 0.9796 | 0.9815 | 0.9815 | 0.9815 |
Framework versions
- Transformers 4.41.1
- Pytorch 2.3.0+cu121
- Datasets 2.19.2
- Tokenizers 0.19.1