search

Evolving-AI-Lab / fooling

Code base for "Deep Neural Networks are Easily Fooled" CVPR 2015 paper
172 stars 59 forks source link

Unable to generate MNIST example #5

Closed antinucleon closed 9 years ago

antinucleon commented 9 years ago

After I change dl_images.hpp, dl_map_elites_images_mnist.cpp and generated MNIST images, I compiled the program. I get segfault. But when I run caffe independently, it works well. I opened GLOG, the output is like

I0525 23:13:19.182551 27365 net.cpp:207] Collecting Learning Rate and Weight Decay.
I0525 23:13:19.182559 27365 net.cpp:130] Network initialization done.
I0525 23:13:19.182565 27365 net.cpp:131] Memory required for data: 0
Segmentation fault (core dumped)
anguyen8 commented 9 years ago

@antinucleon : Were you ever able to generate MNIST images before this segfault?

If you can provide gdb stacktrace , I can help with debugging.

antinucleon commented 9 years ago

The tracking info suggest it is crashed at OpenBlas, I will post full info soon. It is strang that I can run Caffe in CPU mode correctly.

anguyen8 commented 9 years ago

@antinucleon : Ok, make sure that you don't have multiple conflicting versions of Caffe installed on your system.