README.md

metadata

license: other
base_model: nvidia/mit-b0
tags:
  - vision
  - image-segmentation
  - generated_from_trainer
model-index:
  - name: segformer-b0-finetuned-segments-SixrayKnife8-20-2024
    results: []

segformer-b0-finetuned-segments-SixrayKnife8-20-2024

This model is a fine-tuned version of nvidia/mit-b0 on the saad7489/SixraygunTest dataset. It achieves the following results on the evaluation set:

• Loss: 0.2632
• Mean Iou: 0.7518
• Mean Accuracy: 0.8442
• Overall Accuracy: 0.9846
• Accuracy Bkg: 0.9934
• Accuracy Knife: 0.6638
• Accuracy Gun: 0.8755
• Iou Bkg: 0.9864
• Iou Knife: 0.5722
• Iou Gun: 0.6969

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: 6e-05
• train_batch_size: 20
• eval_batch_size: 20
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 50

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Bkg	Accuracy Knife	Accuracy Gun	Iou Bkg	Iou Knife	Iou Gun
0.7462	5.0	20	0.8680	0.5725	0.7955	0.9552	0.9653	0.6150	0.8064	0.9557	0.3394	0.4223
0.5675	10.0	40	0.5259	0.5797	0.6730	0.9685	0.9873	0.3829	0.6486	0.9690	0.3247	0.4455
0.5079	15.0	60	0.4394	0.6394	0.7578	0.9723	0.9859	0.5491	0.7385	0.9731	0.4658	0.4794
0.3976	20.0	80	0.3820	0.6781	0.7446	0.9792	0.9942	0.5443	0.6952	0.9802	0.4938	0.5601
0.3527	25.0	100	0.3454	0.7173	0.8050	0.9816	0.9928	0.6128	0.8094	0.9829	0.5373	0.6318
0.3571	30.0	120	0.3192	0.7336	0.8386	0.9826	0.9917	0.6508	0.8734	0.9843	0.5518	0.6646
0.3201	35.0	140	0.2858	0.7399	0.8390	0.9834	0.9924	0.6540	0.8706	0.9851	0.5637	0.6709
0.3205	40.0	160	0.2774	0.7482	0.8301	0.9846	0.9944	0.6447	0.8512	0.9864	0.5673	0.6911
0.2899	45.0	180	0.2677	0.7497	0.8399	0.9845	0.9937	0.6581	0.8679	0.9864	0.5679	0.6948
0.2672	50.0	200	0.2632	0.7518	0.8442	0.9846	0.9934	0.6638	0.8755	0.9864	0.5722	0.6969

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1