ecc_segformerv3

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

• Loss: 0.1344
• Mean Iou: 0.0005
• Mean Accuracy: 0.0010
• Overall Accuracy: 0.0010
• Accuracy Background: nan
• Accuracy Crack: 0.0010
• Iou Background: 0.0
• Iou Crack: 0.0010

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: 0.0006
• train_batch_size: 1
• eval_batch_size: 1
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• training_steps: 5000

Training results

Training Loss	Epoch	Step	Validation Loss	Mean Iou	Mean Accuracy	Overall Accuracy	Accuracy Background	Accuracy Crack	Iou Background	Iou Crack
0.1306	1.0	1001	0.1114	0.0	0.0	0.0	nan	0.0	0.0	0.0
0.107	2.0	2002	0.1238	0.0000	0.0000	0.0000	nan	0.0000	0.0	0.0000
0.1285	3.0	3003	0.1631	0.0024	0.0049	0.0049	nan	0.0049	0.0	0.0048
0.0887	4.0	4004	0.1083	0.0002	0.0003	0.0003	nan	0.0003	0.0	0.0003
0.0828	5.0	5000	0.1344	0.0005	0.0010	0.0010	nan	0.0010	0.0	0.0010

Framework versions

• Transformers 4.32.0.dev0
• Pytorch 2.0.1+cpu
• Datasets 2.14.4
• Tokenizers 0.13.3