best-model

This model is a fine-tuned version of google/vit-base-patch16-224 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5533
• Accuracy: 0.8289
• Precision: 0.8457
• Recall: 0.8289
• F1: 0.8320
• Precision Indoor: 0.6897
• Recall Indoor: 0.8696
• F1 Indoor: 0.7692
• Support Indoor: 23
• Precision Notapplicable: 0.8182
• Recall Notapplicable: 0.6923
• F1 Notapplicable: 0.75
• Support Notapplicable: 13
• Precision Outdoor: 0.9444
• Recall Outdoor: 0.85
• F1 Outdoor: 0.8947
• Support Outdoor: 40

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: 0.01
• train_batch_size: 16
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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: 0.05
• num_epochs: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1	Precision Indoor	Recall Indoor	F1 Indoor	Support Indoor	Precision Notapplicable	Recall Notapplicable	F1 Notapplicable	Support Notapplicable	Precision Outdoor	Recall Outdoor	F1 Outdoor	Support Outdoor
No log	1.0	19	0.9758	0.7237	0.8166	0.7237	0.7386	0.7059	0.5217	0.6	23	0.4483	1.0	0.6190	13	1.0	0.75	0.8571	40
0.9607	2.0	38	0.5533	0.8289	0.8457	0.8289	0.8320	0.6897	0.8696	0.7692	23	0.8182	0.6923	0.75	13	0.9444	0.85	0.8947	40

Framework versions

• Transformers 4.57.6
• Pytorch 2.9.0+cu126
• Datasets 4.0.0
• Tokenizers 0.22.2