ViT
Collection
ViT For Real Fake Image Classification • 4 items • Updated
vit-real-fake-classification
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:
- eval_loss: 0.1798
- eval_accuracy: 0.9275
- eval_f1: 0.9362
- eval_recall: 0.9649
- eval_precision: 0.9092
- eval_runtime: 110.2958
- eval_samples_per_second: 16.882
- eval_steps_per_second: 0.136
- epoch: 1.9492
- step: 28
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: 128
- eval_batch_size: 128
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 512
- 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
Framework versions
- Transformers 4.40.2
- Pytorch 2.2.1+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 71
Model tree for date3k2/vit-real-fake-classification-v1
Base model
google/vit-base-patch16-224