segformer-finetuned-rwymarkings-2-steps

This model is a fine-tuned version of nvidia/mit-b0 on the Spatiallysaying/rwymarkings dataset. It achieves the following results on the evaluation set:

• Loss: 2.2162
• Mean Iou: 0.0387
• Mean Accuracy: 0.1129
• Overall Accuracy: 0.1050
• Accuracy Backgound : nan
• Accuracy Tdz: 0.0493
• Accuracy Aim: 0.2144
• Accuracy Desig: 0.0922
• Accuracy Rwythr: 0.1765
• Accuracy Thrbar: 0.0140
• Accuracy Disp: 0.2710
• Accuracy Chevron: 0.0023
• Accuracy Arrow: 0.0834
• Iou Backgound : 0.0
• Iou Tdz: 0.0399
• Iou Aim: 0.1158
• Iou Desig: 0.0443
• Iou Rwythr: 0.0980
• Iou Thrbar: 0.0131
• Iou Disp: 0.0266
• Iou Chevron: 0.0020
• Iou Arrow: 0.0085

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: 8
• eval_batch_size: 8
• seed: 1337
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: polynomial
• training_steps: 2

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Backgound	Accuracy Tdz	Accuracy Aim	Accuracy Desig	Accuracy Rwythr	Accuracy Thrbar	Accuracy Disp	Accuracy Chevron	Accuracy Arrow	Iou Backgound	Iou Tdz	Iou Aim	Iou Desig	Iou Rwythr	Iou Thrbar	Iou Disp	Iou Chevron	Iou Arrow
2.2475	0.0455	2	2.2162	0.0387	0.1129	0.1050	nan	0.0493	0.2144	0.0922	0.1765	0.0140	0.2710	0.0023	0.0834	0.0	0.0399	0.1158	0.0443	0.0980	0.0131	0.0266	0.0020	0.0085

Framework versions

• Transformers 4.43.0.dev0
• Pytorch 2.3.0+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1