Terrain-Adaptive Locomotion Skills Using Deep Reinforcement Learning

SIGGRAPH 2016


Reviews

Information

  • Paper topic: Animation and Simulation
  • Software type: Code
  • Able to run a replicability test: False
  • Replicability score: 1
  • Software language: C/C++
  • License: LGPL
  • Build mechanism: premake
  • Dependencies: caffe/boost/opencv/eigen/BulletPhysics/CUDA/Json_cpp/bits/glut/lodepng/
  • Documentation score {0,1,2}: 0
  • Reviewer: Nicolas Mellado <nmellado0@gmail.com>
  • Time spent for the test (build->first run, timeout at 100min): 100min

Source code information

Comments

The provided code comes with many dependencies, which are supposedly handled using premake4.

On windows, I could generate a solution for Visual Studio by calling premake.exe vs2012 from the source code directory (call to more recent versions of VS did not work). I couldn't build the solution, as premake didn't populate the include, build and link directives for the dependencies. I tried to fix it, but I couldn't find a way to properly handle caffe and its dependencies.

I got similar problems on linux, where I could generate a Makefile by calling premake gmake, but I had too many dependencies to fix when calling  make config=debug64.

If you want to contribute with another review, please follow these instructions.

Please consider to cut/paste/edit the raw JSON data attached to this paper.