Vagrant QuantSoftware ToolKit Desktop on Ubuntu 14.x#

This Vagrant box sets up a full QuantSoftware ToolKit Ubuntu installation inside a VirtualMachine with Gnome desktop. It can be both used as a Gnome desktop or through CLI.

DISCLAIMER All credits for QuantSoftware Toolkit goes to the team of QSTK, Here we just build a complete VirtualMachine around it so it can be easily used on 'any' os as long as it supports the free software VirtualBox from Oracle and the opensource tool Vagrant

Also the ipython notebook application is installed for easy usages of python code snippets.

BE PATIENT! It takes a long time to install all the dependency, the good news is you don't need to do anything but wait. No changes are made to your local python installation since everything is run inside a VM

Requirements

• Oracle VM Virtualbox
• Oracle VM VirtualBox Extension Pack
• Vagrantup
• A CPU running OsX, Linux or Windows

Windows Users, Please Note! Using NFS Synced Folders

The current installation uses NFS synced folders (look inside the /Vagrantfile) and is therefor optimised for OsX/Linux hosts. If you like to run this Vagrant box on Microsoft (which is no problem) please read this documentation of Vagrant about NFS Synced Folders and how to handle this on Windows

Quick Start

1. Install Virtualbox, VirtualBox Extension Pack and Vagrantup
2. Clone this repository, and inside run the command:

$ vagrant up

Vagrant will start by installing a vanilla Ubuntu 14.04 inside the VM. It will bring-up the Ubuntu Unity Desktop first and afterwards continue's to install:

1. The python packages: numpy,scipy,matplotlib,pandas,python-dateutil,cvxopt,scikit-learn,pyzmq,ipython,pygments,patsy,statsmodels,lxml,qstk
2. ipython notebook
3. Be patient it takes a long time before it all is completed. (on my slow internet almost 50 minutes) If the full installation was successful you will see the following screen on your terminal:

==> default: Close price of MSFT on 2012/2/15 is :  29.66
==> default: Data looks correct as the close price in default data is 29.66
==> default: 
==> default: Everything works fine: You're all set.
==> default:  

==> default: You find your host ./ directory inside the VM at /vagrant, and it is always in sync
==> default: Use it with your native editor to work on your projects
==> default:  
==> default: Login with the command line: 
==> default: vagrant ssh
==> default:  
==> default: Login on the GUI with:
==> default: username: vagrant
==> default: password: vagrant
==> default: sudo su -    works :-)
==> default:  
==> default: ROOT account ubuntu:
==> default: username: ubuntu
==> default: password: ubuntu
==> default:  
==> default: Happy Coding! :-)

Scroll back the screen and check if no errors occurred. At the end of the installation the ./QSTK/Examples/Validation.py script is executed for testing the installation

Start a terminal on the Linux desktop and type:

ipython notebook 

# or if you like to access notebook from outside the VM:
ipython notebook --ip='*'

use the url http://10.11.12.13:8888/ from your own desktop to reach the VM.

yes the correct ip is 10.11.12.13, VirtualBox will handle the routing :-)

If you like to login the VM through the CLI use:

vagrant ssh

#QSTK Examples & Validation.py script# Can be found in: ./QSTK/QSTK-0.2.8/Examples/

Customise the installation

Feel free to fork me and change :)

All the configuration variables are placed inside the hiera file:

/vagrant/puppet/hiera/hieradata/quantstart-configuration.yaml

To apply new settings just run the script:

$ sudo ./apply_puppet.sh

The Ubuntu desktop tweaks:

/vagrant/puppet/manifests/files/scripts/desktop-setup/desktop-setup.sh

Package List

Below are the different packages used in this Vagrant box:

Python Package List

    - numpy
    - scipy
    - matplotlib
    - pandas
    - python-dateutil
    - cvxopt
    - scikits.statsmodels
    - scikit-learn
    - pyzmq
    - ipython[notebook]
    - pygments
    - patsy
    - statsmodels
    - lxml
    - qstk

Binary Libraries

    - build-essential 
    - python2.7-dev
    - liblapack-dev  
    - libblas-dev 
    - libatlas-base-dev 
    - gfortran 
    - libpng-dev 
    - libjpeg8-dev 
    - libfreetype6-dev 
    - libzmq-dev
    - libxslt-dev
    - quantlib-python
    - python-numpy
    - python-scipy
    - python-matplotlib
    - python-setuptools
    - python-cvxopt
    - unzip

Unity Desktop Apps

    - ubuntu-desktop
    - gnome-terminal
    - firefox
    - chromium-browser
    - vim-gnome
    - geany
    - meld

Enjoy!

Wesseldr, sep-2014