Open JunningWu opened 6 years ago
root@jwy:/home/jwy/DLA3/sw-master/prebuilt/linux# ./nvdla_compiler --prototxt VGG-16/VGG_ILSVRC_16_layers_deploy.prototxt --caffemodel VGG-16/VGG_ILSVRC_16_layers.caffemodel creating new wisdom context... opening wisdom context... parsing caffe network... Marking total 0 outputs (DLA) Error 0x0000000b: Unable to identify outputs for the network: 0 (in ParseTest.cpp, function parseCaffeNetwork(), line 99) (DLA) Error 0x0000000b: (propagating from ParseTest.cpp, function parseAndCompile(), line 164) (DLA) Error 0x0000000b: (propagating from main.cpp, function launchTest(), line 87) root@jwy:/home/jwy/DLA3/sw-master/prebuilt/linux# ./nvdla_compiler --prototxt VGG-16/VGG_ILSVRC_16_layers_deploy.prototxt --caffemodel VGG-16/VGG_ILSVRC_16_layers.caffemodel creating new wisdom context... opening wisdom context... parsing caffe network... Marking total 0 outputs (DLA) Error 0x0000000b: Unable to identify outputs for the network: 0 (in ParseTest.cpp, function parseCaffeNetwork(), line 99) (DLA) Error 0x0000000b: (propagating from ParseTest.cpp, function parseAndCompile(), line 164) (DLA) Error 0x0000000b: (propagating from main.cpp, function launchTest(), line 87) @JunningWu @prasshantg @jwise
i had the same error when compiling vgg16 as JIAWENYANG faced. Can nvdla support vgg16?
the prototxt and caffemodel is downloaded from this site: http://www.robots.ox.ac.uk/%7Evgg/research/very_deep/
when I compile the VGG16 CNN, there is an error:
when debug this using gdb, it shows like this:
0 0x000000000040b34c in parseCaffeNetwork(TestAppArgs const, TestInfo) ()
1 0x000000000040b984 in parseAndCompile(TestAppArgs const, TestInfo) ()
2 0x0000000000406abd in launchTest(TestAppArgs const*) ()
3 0x0000000000406eea in main ()
since there does not have some NEW features apart from Alexnet or Lenet, this error is very strange.