Segnet Caffe Upsample Top index 0 out of Range-check scale settings match input pooling layer ' s downsample setup

I0305 the: -:08.603567 14672Solver.cpp: -] Solving vgg_ilsvrc_16_layeri0305 the: -:08.603581 14672Solver.cpp:251] Learning rate policy:stepf0305 -:xx:32.416029 14672Upsample_layer.cpp:127] Upsample Top Index0  outof Range-check scale settings match input pooling layer's downsample SetupCheck failure Stack Trace: * * *@ 0X7F372F8AA5CDgoogle::logmessage::fail () @0x7f372f8ac433Google::logmessage::sendtolog () @0x7f372f8aa15bGoogle::logmessage::flush () @0x7f372f8ace1egoogle::logmessagefatal::~logmessagefatal () @0x7f373010616fCaffe::upsamplelayer<>:: Backward_cpu () @0x7f37300c8042Caffe::net<>:: Backwardfromto () @0x7f37300c8241Caffe::net<>:: Backward () @0x7f37301efc7dCaffe::solver<>:: Step () @0x7f37301f0754Caffe::solver<>:: Solve () @0x408019train () @0x405ce8main@0x7f372ef5e830__libc_start_main@0x406289_start@ (nil) (unknown) aborted (core dumped)

This is a CPU version error, and I see the workaround as follows:

Since this is the author's own Caffe, and the version is older, so you can make the following changes:

Comment out the line 127,128 in Upsample_layer.cpp directly:

And then

Make Cleanmake All

That is, recompile Caffe, because this CPP is Caffe code, so you have to recompile caffe.

Segnet Caffe Upsample Top index 0 out of Range-check scale settings match input pooling layer ' s downsample setup