Feedforward Closedloop Learning (FCL)
Project description
Feedforward closed loop learning (FCL) is a learning algorithm which adds flexibility to autonomous agents.
A designer defines an initial behaviour as a reflex and then FCL learns from the reflex to develop new flexible behaviour.
The Python documentation can be obtained with:
import feedforward_closedloop_learning as fcl help(fcl)
The Python API is identical to the C++ API: The header files fcl.h, neuron.h and layer.h contain docstrings for all important calls. The doxygen generated documentation can be found here: https://github.com/glasgowneuro/feedforward_closedloop_learning/tree/master/docs
The best way to get started is to look at the script in tests_py: https://github.com/glasgowneuro/feedforward_closedloop_learning/tree/master/tests_py
A full application using the Python API is our vizdoom agent: https://github.com/glasgowneuro/fcl_demos
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Hashes for feedforward_closedloop_learning-1.2.3.tar.gz
Algorithm | Hash digest | |
---|---|---|
SHA256 | 120003541bfc5ab361741759a6d3d77ab0f3a03affa6bc6867857fd2b7a29646 |
|
MD5 | f5278d601676aa8732d025ea68e3ab31 |
|
BLAKE2b-256 | 3750fd36a09088325ac066ba818c47c32b3977e3f11f21e5d9f8e7f80fa2291e |