Skip to content

Introduction

DavidRagazzi edited this page Nov 17, 2014 · 9 revisions

NuPIC Studio is an powerful all-in-one toolkit allowing users to create an HTM neural network from scratch, train it, collect statistics, and share it with the members of the community. It is not just a visualization tool, but an HTM builder, debugger and laboratory for experiments. It's ideal for newbies with little intimate knowledge of NuPIC code, as well as experts that desire better productivity.

Among its features and advantages:

  • Users can open, save, and change their "HTM projects" and those of other developers. A typical project contains data to be trained, neural network configuration, statistics, etc, which can be shared in order to be analyzed or integrated with other projects.
  • The HTM engine is the original NuPIC library (Python distribution). This means no port, no bindings, no re-implementation, etc. So any changes in the original NuPIC source can be immediately viewed. This helps users who wish to test improvements, like new encoders or even hierarchy, attention, and motor integration.

This tutorial aims to provide among other things: an overview of the main features of NuPIC Studio, a guide on how to create your own projects, and a detailed explanation of how the inputs are organized and stored, and finally how to create your own encoders and integrate them into your project.

Clone this wiki locally