search

advaitsave / Multiclass-Semantic-Segmentation-CamVid

Tensorflow 2 implementation of complete pipeline for multiclass image semantic segmentation using UNet, SegNet and FCN32 architectures on Cambridge-driving Labeled Video Database (CamVid) dataset.
82 stars 41 forks source link

请问训练模型的时候的IoU分数是从哪儿来的？ #11

Open Shirvin opened 4 years ago

Shirvin commented 4 years ago

请问训练模型的时候的IoU分数是从哪儿来的？