search

xmengli / H-DenseUNet

TMI 2018. H-DenseUNet: Hybrid Densely Connected UNet for Liver and Tumor Segmentation from CT Volumes
526 stars 161 forks source link

a question about train data and result #73

Open Hung-Hsiang-617 opened 4 years ago

Hung-Hsiang-617 commented 4 years ago

Do training dataests need to use Resnet method to get coarse segmentation before using h-denseunet? Because I directly input training data and the training effect is very poor. Can you provide an estimate of how many epochs are trained to have better results?

xmengli commented 4 years ago

It is not necessary to train Resnet, since the coarse segmentation is only used in the inference stage.

Training takes around 3 days using 4 12GB TITANXP GPU.