wmc-wmk811-v0-vit-special_map_det_0917

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

• Loss: 0.0354
• Accuracy: 0.9882
• Precision: 0.9872
• Recall: 0.9854
• F1: 0.9863

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: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy	Precision	Recall	F1
0.0471	0.2158	400	0.0651	0.9766	0.9793	0.9662	0.9724
0.0664	0.4317	800	0.0445	0.9874	0.9879	0.9828	0.9853
0.0391	0.6475	1200	0.0476	0.9833	0.9826	0.9785	0.9805
0.0478	0.8633	1600	0.0354	0.9882	0.9872	0.9854	0.9863

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.0+cu121
• Datasets 3.0.0
• Tokenizers 0.19.1