vit-large-patch16-224-in21k-testing-dungeons-lora-27Dec24-0001

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

• Loss: 0.1076
• Accuracy: 0.9607

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: 0.005
• 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: 12
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.7273	2	1.5502	0.2946
No log	1.7273	4	1.1408	0.6696
No log	2.7273	6	1.0113	0.6036
No log	3.7273	8	0.6030	0.8411
5.1081	4.7273	10	0.4665	0.8625
5.1081	5.7273	12	0.4145	0.8643
5.1081	6.7273	14	0.2846	0.9107
5.1081	7.7273	16	0.2386	0.9125
5.1081	8.7273	18	0.1564	0.9554
0.7653	9.7273	20	0.1178	0.9679
0.7653	10.7273	22	0.1241	0.9536
0.7653	11.7273	24	0.1076	0.9607

Framework versions

• PEFT 0.14.0
• Transformers 4.47.1
• Pytorch 2.5.1+cu121
• Datasets 3.2.0
• Tokenizers 0.21.0