segformer-b5-finetuned-ce-head-batch2

This model was trained from scratch on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.0575
• Mean Iou: 0.7701
• Mean Accuracy: 0.8473
• Overall Accuracy: 0.9774
• Accuracy Bg: 0.9891
• Accuracy Head: 0.7054
• Iou Bg: 0.9768
• Iou Head: 0.5634

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: 3e-05
• train_batch_size: 2
• eval_batch_size: 2
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Bg	Accuracy Head	Iou Bg	Iou Head
0.0192	2.9412	100	0.0164	0.8513	0.9095	0.9934	0.9968	0.8221	0.9933	0.7094
0.0071	5.8824	200	0.0197	0.8495	0.8819	0.9933	0.9982	0.7656	0.9932	0.7058
0.0345	8.8235	300	0.0154	0.8745	0.9312	0.9940	0.9968	0.8657	0.9939	0.7551
0.0056	11.7647	400	0.0158	0.8668	0.9375	0.9937	0.9961	0.8789	0.9936	0.7401
0.0217	14.7059	500	0.0149	0.8736	0.9306	0.9936	0.9966	0.8646	0.9934	0.7538
0.0924	17.6471	600	0.0133	0.8738	0.9438	0.9943	0.9963	0.8913	0.9942	0.7534

Framework versions

• Transformers 4.46.2
• Pytorch 2.5.1
• Datasets 3.1.0
• Tokenizers 0.20.3