SegFormer_mit-b5_Final-Set4-Grayscale_Test_4_lr0.0001

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

• Loss: 0.1193
• Mean Iou: 0.8167
• Mean Accuracy: 0.8517
• Overall Accuracy: 0.9601
• Accuracy Background: 0.9780
• Accuracy Melt: 0.5944
• Accuracy Substrate: 0.9827
• Iou Background: 0.9616
• Iou Melt: 0.5614
• Iou Substrate: 0.9270

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: 0.0001
• train_batch_size: 4
• eval_batch_size: 4
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• lr_scheduler_warmup_steps: 100
• num_epochs: 1
• mixed_precision_training: Native AMP

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Melt	Accuracy Substrate	Iou Background	Iou Melt	Iou Substrate
0.1133	0.8850	50	0.1193	0.8167	0.8517	0.9601	0.9780	0.5944	0.9827	0.9616	0.5614	0.9270

Framework versions

• Transformers 4.41.2
• Pytorch 2.0.1+cu117
• Datasets 2.19.2
• Tokenizers 0.19.1