search

IBBM / Cascaded-FCN

Source code for the MICCAI 2016 Paper "Automatic Liver and Lesion Segmentation in CT Using Cascaded Fully Convolutional NeuralNetworks and 3D Conditional Random Fields"
Other
304 stars 127 forks source link

Class Weight Selection #24

Closed saqibqamar closed 6 years ago

saqibqamar commented 7 years ago

why use class weight 1, 13,120 ? Can we use other number for weights ?...On what basis you are using this weights..

class_loss_weights: 1 class_loss_weights: 13 class_loss_weights: 120

mohamed-ezz commented 6 years ago

Class weights were picked so that weight*pixel_frequency for each class is the same (up to slight approximation)