parking-utcustom-train-SF-RGB-b0_3

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: 0.6640
• Mean Iou: 0.3046
• Mean Accuracy: 0.9138
• Overall Accuracy: 0.9138
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.9138
• Iou Unlabeled: 0.0
• Iou Parking: 0.0
• Iou Unparking: 0.9138

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: 3.5e-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
1.0744	20.0	20	1.0527	0.2035	0.6104	0.6104	nan	nan	0.6104	0.0	0.0	0.6104
0.898	40.0	40	0.9525	0.2756	0.8267	0.8267	nan	nan	0.8267	0.0	0.0	0.8267
0.7959	60.0	60	0.8295	0.2804	0.8413	0.8413	nan	nan	0.8413	0.0	0.0	0.8413
0.7014	80.0	80	0.7152	0.3014	0.9041	0.9041	nan	nan	0.9041	0.0	0.0	0.9041
0.6723	100.0	100	0.6346	0.3125	0.9374	0.9374	nan	nan	0.9374	0.0	0.0	0.9374
0.6464	120.0	120	0.6640	0.3046	0.9138	0.9138	nan	nan	0.9138	0.0	0.0	0.9138

Framework versions

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