search

0bserver07 / One-Hundred-Layers-Tiramisu

Keras Implementation of The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation by (Simon Jégou, Michal Drozdzal, David Vazquez, Adriana Romero, Yoshua Bengio)
https://arxiv.org/abs/1611.09326
MIT License
197 stars 54 forks source link

HeUniform in Keras? #3

Closed 0bserver07 closed 7 years ago

0bserver07 commented 7 years ago

init="he_uniform"