parking-utcustom-train-SF-RGB-b0_6

This model is a fine-tuned version of nvidia/mit-b0 on the sam1120/parking-utcustom-train dataset. It achieves the following results on the evaluation set:

• Loss: 0.6930
• Mean Iou: 0.2916
• Mean Accuracy: 0.8749
• Overall Accuracy: 0.8749
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.8749
• Iou Unlabeled: 0.0
• Iou Parking: 0.0
• Iou Unparking: 0.8749

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: 4.5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.05
• num_epochs: 120

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Unlabeled	Accuracy Parking	Accuracy Unparking	Iou Unlabeled	Iou Parking	Iou Unparking
1.0679	20.0	20	1.0841	0.2124	0.6371	0.6371	nan	nan	0.6371	0.0	0.0	0.6371
0.8545	40.0	40	1.0206	0.2672	0.8016	0.8016	nan	nan	0.8016	0.0	0.0	0.8016
0.7543	60.0	60	0.8816	0.2796	0.8389	0.8389	nan	nan	0.8389	0.0	0.0	0.8389
0.6537	80.0	80	0.7175	0.2924	0.8771	0.8771	nan	nan	0.8771	0.0	0.0	0.8771
0.6271	100.0	100	0.6655	0.2991	0.8972	0.8972	nan	nan	0.8972	0.0	0.0	0.8972
0.6034	120.0	120	0.6930	0.2916	0.8749	0.8749	nan	nan	0.8749	0.0	0.0	0.8749

Framework versions

• Transformers 4.30.2
• Pytorch 2.0.1+cu117
• Datasets 2.13.1
• Tokenizers 0.13.3