parking-utcustom-train-SF-RGB-b0_5

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: 1.0414
• Mean Iou: 0.2695
• Mean Accuracy: 0.8085
• Overall Accuracy: 0.8085
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.8085
• Iou Unlabeled: 0.0
• Iou Parking: 0.0
• Iou Unparking: 0.8085

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: 4.25e-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
1.1636	20.0	20	1.1342	0.0522	0.1567	0.1567	nan	nan	0.1567	0.0	0.0	0.1567
0.9787	40.0	40	1.1195	0.2039	0.6116	0.6116	nan	nan	0.6116	0.0	0.0	0.6116
0.887	60.0	60	1.0823	0.2420	0.7259	0.7259	nan	nan	0.7259	0.0	0.0	0.7259
0.7959	80.0	80	0.9519	0.2693	0.8080	0.8080	nan	nan	0.8080	0.0	0.0	0.8080
0.7344	100.0	100	0.8902	0.2827	0.8481	0.8481	nan	nan	0.8481	0.0	0.0	0.8481
0.7391	120.0	120	1.0414	0.2695	0.8085	0.8085	nan	nan	0.8085	0.0	0.0	0.8085

Framework versions

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