parking-utcustom-train-SF-RGB-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: 0.3336
• Mean Iou: 0.3310
• Mean Accuracy: 0.9930
• Overall Accuracy: 0.9930
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.9930
• Iou Unlabeled: 0.0
• Iou Parking: 0.0
• Iou Unparking: 0.9930

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: 1e-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
0.7453	20.0	20	0.7553	0.3187	0.9562	0.9562	nan	nan	0.9562	0.0	0.0	0.9562
0.5909	40.0	40	0.5185	0.3316	0.9948	0.9948	nan	nan	0.9948	0.0	0.0	0.9948
0.473	60.0	60	0.3947	0.3327	0.9982	0.9982	nan	nan	0.9982	0.0	0.0	0.9982
0.4101	80.0	80	0.3458	0.3325	0.9975	0.9975	nan	nan	0.9975	0.0	0.0	0.9975
0.3731	100.0	100	0.3418	0.3315	0.9945	0.9945	nan	nan	0.9945	0.0	0.0	0.9945
0.3575	120.0	120	0.3336	0.3310	0.9930	0.9930	nan	nan	0.9930	0.0	0.0	0.9930

Framework versions

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