vit-large-patch16-224-in21k-dungeon-geo-morphs-denoised-04Dec24-002

This model is a fine-tuned version of google/vit-large-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.1107
• Accuracy: 0.9657

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 32
• optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 40
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
1.4876	4.0	10	1.1611	0.6545
0.6201	8.0	20	0.5442	0.9152
0.1543	12.0	30	0.2724	0.9556
0.0344	16.0	40	0.1593	0.9636
0.0095	20.0	50	0.1314	0.9657
0.0047	24.0	60	0.1091	0.9657
0.0033	28.0	70	0.1139	0.9636
0.0029	32.0	80	0.1107	0.9657

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1+cu121
• Datasets 3.1.0
• Tokenizers 0.20.3