vit-large-ai-or-not

This model is a fine-tuned version of google/vit-large-patch16-224 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1039
• Accuracy: 0.9581

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: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 2
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
0.3925	0.1808	200	0.4045	0.8786
0.2803	0.3617	400	0.2386	0.9044
0.2235	0.5425	600	0.1893	0.9173
0.217	0.7233	800	0.1597	0.9398
0.1865	0.9042	1000	0.1413	0.9420
0.1309	1.0850	1200	0.1474	0.9517
0.1008	1.2658	1400	0.1914	0.9420
0.0793	1.4467	1600	0.1557	0.9441
0.0804	1.6275	1800	0.2301	0.9313
0.0814	1.8083	2000	0.1039	0.9581
0.0446	1.9892	2200	0.1124	0.9635

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 3.0.2
• Tokenizers 0.19.1