parking-utcustom-train-SF-RGB-b0_4

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: 0.6504
• Mean Iou: 0.3176
• Mean Accuracy: 0.9528
• Overall Accuracy: 0.9528
• Accuracy Unlabeled: nan
• Accuracy Parking: nan
• Accuracy Unparking: 0.9528
• Iou Unlabeled: 0.0
• Iou Parking: 0.0
• Iou Unparking: 0.9528

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: 4e-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
1.1564	20.0	20	1.1510	0.1413	0.4239	0.4239	nan	nan	0.4239	0.0	0.0	0.4239
0.924	40.0	40	1.1282	0.2675	0.8024	0.8024	nan	nan	0.8024	0.0	0.0	0.8024
0.8182	60.0	60	0.9038	0.2908	0.8723	0.8723	nan	nan	0.8723	0.0	0.0	0.8723
0.6976	80.0	80	0.7238	0.3070	0.9211	0.9211	nan	nan	0.9211	0.0	0.0	0.9211
0.6843	100.0	100	0.6362	0.3221	0.9663	0.9663	nan	nan	0.9663	0.0	0.0	0.9663
0.6309	120.0	120	0.6504	0.3176	0.9528	0.9528	nan	nan	0.9528	0.0	0.0	0.9528

Framework versions

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