search

xwjabc / hed

A PyTorch reimplementation of Holistically-Nested Edge Detection
170 stars 42 forks source link

Training #37

Open long123524 opened 3 years ago

long123524 commented 3 years ago

image I used my own data set, but the result was a black image with all 0.5 values. How should I improve it?

xwjabc commented 3 years ago

I would suggest following https://github.com/xwjabc/hed/issues/36 and attempt to load the pre-trained VGG weights except the first layer.