Skip to content

README.md

Latest commit

 

History

History
15 lines (11 loc) · 883 Bytes

README.md

File metadata and controls

15 lines (11 loc) · 883 Bytes

A very basic implementation of Reservoir that can learn chaotic systems

To run you need to have the following installed in your working python environment:

  1. Numpy
  2. Jitcode
  3. NetworkX

Example(Rossler Attractor):

  1. To run: python3 lorenz.py

  2. --> Command line arguments can be given. For complete list of arguments, see lorenz.py

  3. --> Warning: Parameters, leaky_rate and inputScaling_radius are quite sensitive for reasons currently unknown to me. Avoid changing them!

  4. --> At the moment, rossler is not behaving normally. Since every compilation produces different random numbers for weights, sometimes the reservoir output blows up and sometimes it learns the system quite well. Parameters for rossler at not great right now and its better to avoid it.