parking-utcustom-train-SF-RGB-b0_1

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: 0.7736
• Mean Iou: 0.3038
• Mean Accuracy: 0.9115
• Overall Accuracy: 0.9115
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.9115
• Iou Unlabeled: 0.0
• Iou Parking: 0.0
• Iou Unparking: 0.9115

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: 2.5e-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
1.0406	20.0	20	1.0478	0.2219	0.6657	0.6657	nan	nan	0.6657	0.0	0.0	0.6657
0.9316	40.0	40	0.9617	0.2595	0.7786	0.7786	nan	nan	0.7786	0.0	0.0	0.7786
0.8588	60.0	60	0.8796	0.2921	0.8762	0.8762	nan	nan	0.8762	0.0	0.0	0.8762
0.7749	80.0	80	0.8291	0.2977	0.8930	0.8930	nan	nan	0.8930	0.0	0.0	0.8930
0.7452	100.0	100	0.7801	0.3075	0.9226	0.9226	nan	nan	0.9226	0.0	0.0	0.9226
0.7238	120.0	120	0.7736	0.3038	0.9115	0.9115	nan	nan	0.9115	0.0	0.0	0.9115

Framework versions

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