Recent content by amewo

I read this publication (in attachment). And according to the author, he gets a very good prediction using neural network and wavelet transform. P.S. Reference to the work I've found on some n page of this thread, I would like to insert a link to the post, but could not find it, so I spread the...

By the way! Here's a link to the publication of the person: NeuroEvolution of Augmenting Topologies Basic for me: http://nn.cs.utexas.edu/keyword?stanley:phd04 I use his ideas as a basis for the development of GA. Now I'm trying to understand the wavelet transform, when I have free time.

states.pb.h and states.pb.cpp You can get from file states.proto using the library protobuf: https://code.google.com/p/protobuf/ simple example on my os linux: protoc --cpp_out=./ states.proto In the file stats.proto I described the various states of different objects, it is I need to save the...

Thank you for your post. I want to implement their own neural network, because then I'll know exactly how it works, and I think I can write more optimal code that will be faster than existing solutions. I already had experience with neural networks a couple of years ago and tried to do a...

My repository with neuro_net project https://github.com/amewo/neuro_net

Hello everyone! I want to deal with neural networks, and now I write implementations on C++. I want to use a genetic algorithm for the selection of weights and the structure of the neural network, and if it will be successfull, then also realize it on nvidia cuda. Thread is not dead? PS. Sorry...