re: You could look at examples of its usage in mlpack and that may help.Thank you for your advices, maybe I am a freshman and not familiar with Armadillo. From its document introduction, it is a good choice for scientific computing. I will learn more about this tool.
There is another question for build the mlpack. The README page said that " You can specify individual components which you want to build", but I can not find build the reinforcement learning components. Did I miss something or there is no shortcut for this component? If there is no such component, how can I add this shortcut in CMakeList.txt. Thanks, Xiaohong At 2019-01-13 00:44:56, "Ryan Curtin" <[email protected]> wrote: >On Sat, Jan 12, 2019 at 12:13:11PM +0800, problemset wrote: >> Hi, Ryan, >> Due to my schedule, there may some time delay for the PER project >> (https://github.com/mlpack/mlpack/pull/1614). Now, I have more time >> devote to this project. I have implement the basic skeleton for the >> PER and will add unit test to do sanity check and function utility >> check. > >No worries, that's just fine that there was a delay. > >> So, here are some questions (sorry for bother you again): >> 1) All the unit test are placed into the test folder and use the boost >> unit test suit? > >Yes, these should be just like the other tests in src/mlpack/tests/. > >> 2) Can we use vector in STL rather than Armadillo, I found the >> Armadillo document is a little difficult to use for me as reference. >> Or is there any equivalent reference book/materials ? > >I've never heard that the Armadillo documentation is hard to understand >and I've always thought it was pretty simple. Basically it tries to >emulate MATLAB or Octave notation as much as possible. You could look >at examples of its usage in mlpack and that may help. > >But in any case, we should be using Armadillo for linear algebra and not >STL vectors, which are not optimized for that. This can make a big >runtime difference. > >-- >Ryan Curtin | "I like fish sticks." >[email protected] | - Benny
_______________________________________________ mlpack mailing list [email protected] http://knife.lugatgt.org/cgi-bin/mailman/listinfo/mlpack
