vit-large-patch16-224-dungeon-geo-morphs-0-4-28Nov24-006

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

• Loss: 0.0747
• Accuracy: 0.9821

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: 1e-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
1.6194	4.0	10	1.2322	0.6214
0.7978	8.0	20	0.5919	0.925
0.2576	12.0	30	0.2721	0.9679
0.0723	16.0	40	0.1548	0.9786
0.0202	20.0	50	0.1066	0.9768
0.0067	24.0	60	0.0747	0.9821
0.0035	28.0	70	0.0754	0.9768
0.0027	32.0	80	0.0730	0.9786

Framework versions

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