Skip to content

Latest commit

 

History

History
91 lines (54 loc) · 2.28 KB

HOWTO.md

File metadata and controls

91 lines (54 loc) · 2.28 KB
Raw

How to run this code

Prerequisites:

This code depends on:

  • vowpal wabbit (aka vw)
  • R
  • ggplot2 (an R library to create charts)
  • GNU make
  • git (to clone this repository)
  • bash, perl, and python (these are usually preinstalled and available on all Linux and MacOs systems)

Installation of prerequisites:

Linux: Ubuntu, Mint, or any Debian derivative

sudo apt-get install make vowpal-wabbit r-base r-base-core r-cran-ggplot2 git

Other Linux systems

Packages are usually named differently. Contributions to this section very welcome

MacOs / OS-X

Use brew to install the above packages Contributions to this section very welcome

Windows

The only sane way to run this code in a Windows environment, is to install run Ubuntu Linux on a VM (virtual machine) inside Windows, and use the Ubuntu instructions in the VM.

For instructions how to set up a VM on Windows, follow these youtube videos:

One you have Ubuntu on Windows you just install all the prerequisites. e.g. in a terminal:

sudo apt-get install make r-base r-base-core r-cran-ggplot2 vowpal-wabbit git

inside it, to run everything from start to finish.

Running the code

Using git, you clone this repository:

git clone https://github.com/arielf/weight-loss

And change directory to it:

cd weight-loss

Finally type:

make

It should produce a file scores.txt with your weight-loss scores.

To get a chart of the scores:

make sc

or

make score-chart

Changing make parameters

There are a few adjustable variables (which have reasonable defaults) in the Makefile, and which you may change if interested:

To change these you may call make with arguments changing the values, like this:

make VarName1=Value1 VarName2=Value2 ...

The current variables and their defaults settings are:

BS = 7            # -- bootstrapping rounds
P = 4             # -- multiple passes over the data
L = 0.05          # -- learning rate
L2 = 1.85201e-08  # -- L2 regularization
NDAYS = 3         # -- Aggregate consecutive daily-data up to NDAYS days