Bert Peters
99cf0740fe
Now all n individual images of w×h are packaged as a w×(n×h) image, with the texture coordinates properly mapped so the individual rectangles still get the proper texture. This optimization causes slight glitches around the edges of the tiles, but is so much faster that I think it's worth it.
Functional MRI for Neural Networks
An attempt to create real-time in depth visualisation for neural networks.
Requirements
The following programs/libraries need to be installed on your system:
- CMake (at least 3.1)
- Caffe
- Google log (glog)
- OpenCV
- Boost System
- PNG++
Note that most of these are dependencies of Caffe, and thus will probably already be on your system.
Building
The build process is based on CMake. This means you can easily create a build as follows. Starting from the project root:
mkdir build && cd build
cmake ..
make
Compilation is a little slow due to the inclusion of Boost.