Contributing¶
Thank you for your interest in helping nupic.torch improve and evolve. Numenta maintains and runs this code to support ongoing research and exploration of biologically inspired ideas applied to Deep Networks.
Before your pull requests will be reviewed by our team of committers, you’ll need to sign our Contributor License.
Before considering contribution, we hope you’ll try to understand the biology behind these theories. It would be a good idea to become familiar with HTM Theory and read Why Neurons Have Thousands of Synapses, A Theory of Sequence Memory in Neocortex.
File Bugs¶
If you found a bug, please report it in as much detail as possible. Copy / paste of console text is better than screenshots of console logs.
Pull Requests¶
Pull Requests should provide unit tests and address an issue.
Our Community¶
There are many people interested Hierarchical Temporal Memory. You can find them on HTM Forum. Please post questions about nupic.torch
in the Machine Learning subforum.
Committers¶
ml-committers
may merge Pull Requests.
Release Process¶
See our release process at RELEASE.md.
Coding Standards and Guides¶
Run flake8
before creating a PR¶
To run Flake8, install requirements-dev.txt
before invoking flake8
. This enforces standard PEP8, and runs to validate all pull requests.
Create Unit tests¶
See examples in tests.