Maria Sargsyan

@hwangdeyu After i turned off the back-to-back optimizer in the beginning of init file, I am printing a message both at the beginning of the _optimize_conv_batchnorm_fusion(g, node, consumer_nodes) and in...

i guess the problem might be not in the fusing but in the type of the batchnormalization layer used which is SyncBatchNormalization. I have prepared a minimal example for it....

@hwangdeyu Can you tell just for my understanding what happens when the operation that is present in the saved model but not present in the onnx operations? Is this operation...

Hi @ctensmeyer , Is there a possibility to use other layer or maybe combination of other layers in Caffe to replace BilinearInterpolation Layer (maybe deconv or by bilinear filter) ?...

Yes, I feel that the current implementation is also wrong, can someone give update on this?