parking-utcustom-train-SF-RGBD-b5_1

This model is a fine-tuned version of nvidia/mit-b5 on the sam1120/parking-utcustom-train dataset. It achieves the following results on the evaluation set:

• Loss: 0.0476
• Mean Iou: 0.4942
• Mean Accuracy: 0.9883
• Overall Accuracy: 0.9883
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.9883
• Iou Unlabeled: nan
• Iou Parking: 0.0
• Iou Unparking: 0.9883

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: 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: 150

Training results

Training Loss	Epoch	Step	Accuracy Parking	Accuracy Unlabeled	Accuracy Unparking	Iou Parking	Iou Unlabeled	Iou Unparking	Validation Loss	Mean Accuracy	Mean Iou	Overall Accuracy
0.4573	20.0	20	nan	nan	0.9829	0.0	0.0	0.9829	0.3024	0.9829	0.3276	0.9829
0.2183	40.0	40	nan	nan	0.9953	0.0	0.0	0.9953	0.2365	0.9953	0.3318	0.9953
0.1266	60.0	60	nan	nan	1.0	nan	nan	1.0	0.0999	1.0	1.0	1.0
0.0929	80.0	80	nan	nan	0.9972	0.0	nan	0.9972	0.0590	0.9972	0.4986	0.9972
0.0649	100.0	100	0.0346	0.4992	0.9984	0.9984	nan	nan	0.9984	nan	0.0	0.9984
0.0537	120.0	120	0.0377	0.4980	0.9960	0.9960	nan	nan	0.9960	nan	0.0	0.9960
0.0536	140.0	140	0.0476	0.4942	0.9883	0.9883	nan	nan	0.9883	nan	0.0	0.9883

Framework versions

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