Skip to content

Adapt pending predefined layers

-Updated pom.xml -Updated layer templates for Input and Output -Updated layer templates for Flatten, FullyConnected, GlobalPooling and Lrn -Updated convolution and pooling layer templates -Added text for currently unsupported layers -Updated softmax layer -Corrected dropout and output layer (include sigmoid). Added learning_rate_decay again since it is indeed supported in Caffe2

Merge request reports