README.md

metadata

license: other
base_model: nvidia/mit-b0
tags:
  - vision
  - image-segmentation
  - generated_from_trainer
model-index:
  - name: segformer-b0-finetuned-segments-SixrayKnife8-21-2024_saad1
    results: []

segformer-b0-finetuned-segments-SixrayKnife8-21-2024_saad1

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: 0.1146
• Mean Iou: 0.8471
• Mean Accuracy: 0.9142
• Overall Accuracy: 0.9899
• Accuracy Bkg: 0.9952
• Accuracy Knife: 0.8462
• Accuracy Gun: 0.9012
• Iou Bkg: 0.9906
• Iou Knife: 0.7813
• Iou Gun: 0.7695

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
0.2793	5.0	20	0.2864	0.8107	0.9046	0.9864	0.9921	0.8291	0.8925	0.9869	0.7282	0.7169
0.2448	10.0	40	0.2176	0.8159	0.9001	0.9871	0.9932	0.8241	0.8829	0.9876	0.7319	0.7284
0.2061	15.0	60	0.1960	0.8225	0.9093	0.9875	0.9930	0.8324	0.9024	0.9881	0.7476	0.7317
0.1731	20.0	80	0.1698	0.8291	0.8991	0.9884	0.9947	0.8120	0.8907	0.9890	0.7555	0.7428
0.1513	25.0	100	0.1435	0.8371	0.8993	0.9891	0.9954	0.8245	0.8780	0.9897	0.7643	0.7574
0.1401	30.0	120	0.1334	0.8400	0.9112	0.9893	0.9947	0.8399	0.8990	0.9899	0.7720	0.7582
0.1359	35.0	140	0.1222	0.8449	0.9050	0.9898	0.9957	0.8335	0.8859	0.9904	0.7753	0.7691
0.1498	40.0	160	0.1196	0.8460	0.9092	0.9898	0.9955	0.8367	0.8955	0.9905	0.7780	0.7696
0.1255	45.0	180	0.1160	0.8475	0.9109	0.9899	0.9955	0.8423	0.8948	0.9906	0.7810	0.7710
0.1247	50.0	200	0.1146	0.8471	0.9142	0.9899	0.9952	0.8462	0.9012	0.9906	0.7813	0.7695

Framework versions

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