README

This README explains how to get NAB up and running on your machine. NAB is a novel benchmark for evaluating algorithms for anomaly detection in streaming, 
real-time applications. It is comprised of over 50 labeled real-world and artificial timeseries data files plus a novel scoring mechanism designed for 
real-time applications.

In order to install NAB on you local machine you can use NAB docker file provided here. It's been built on top of another docker from "numenta/nupic". 
To run a docker file you need to first, install docker on you machine and then build the current file so it is ready to run. 

To install docker you need to follow instruction on: 

https://docs.docker.com/engine/installation/

When done by installing docker, you need to buil this docker using:

docker build -t you_docker_name .

Now you can run the docker and NAB will be there for you:

docker run your_docker_name


You can find full explanation about NAB and how to install it on:

https://github.com/numenta/NAB

But to have a quick start and see some results you can run the following command:

cd /path/to/nab
python run.py -d numenta --detect --score --normalize

Good Luck

NAB-Docker

# Set the base image to Ubuntu
FROM numenta/nupic

# File Author / Maintainer
MAINTAINER Alireza Hajibagheri ([email protected])

# Set enviroment variables needed by NuPIC
ENV NUPIC /usr/local/src/NAB

# OPF needs this
ENV USER docker

# Change work directory
WORKDIR /usr/local/src

# Clone NAB
RUN git clone https://github.com/numenta/NAB

# Change directory to NAB
WORKDIR /usr/local/src/NAB

# Install NAB python requirments
RUN pip install -r requirements.txt

# Install NAB
RUN python setup.py install --user

WORKDIR /home/docker

Thanks for doing this! If you don't mind, I'd like to add this to https://github.com/nupic-community/NAB-Dockerfile