mvdbosch/AtosCodexDemo

Code used for the Atos Codex Data Scientist Workbench (DSW) demo session, based on the Atos Codex Fabric platform

Atos Codex Data Scientist Workbench Demo

Code used for the Atos Codex Data Scientist Workbench (DSW) demonstration

Instructions

Clone this repository on a CentOS Linux environment, which has the blueprint of Atos Codex 2.3 Fabric or PaaS installed. (Note: It is possible to execute this demo on a different environment. However, this would require a manual component setup, similar to the Atos Codex PaaS blueprint)

Target location should be in the user's document directory, e.g.: /home/marcel/Documents/

Use the following command:

git clone https://www.github.com/mvdbosch/AtosCodexDemo.git

Requirements

The demo requires the following CentOS / YUM packages to be installed on the operating system.

yum install libpng-devel libjpeg-devel libxml2-devel geos-devel

The R code is requiring the following R libraries to be installed. This can be done by the following command:

install.packages(c('shiny','data.table','ggplot2','ggmap','grid','gridExtra','stringr','XML','pmml'))

The following Python packages are required. You can install them using pip or pip3 (for Python3)

pip3 install pandas numpy geopandas matplotlib  ipywidgets IPython sklearn scipy scikit-learn 

High-level demo storyline

1. Perform git clone from github page
2. Create new Talend project + Click on Jobs, import the components from the "Talend ETL code" directory. + Execute main
3. Open KNIME and Import KNIME Workflow. Reset + Execute
4. Rstudio : Step by step walkthrough of "ExploreAndVisualize.R"
5. Rstudio : Full execution of the Shiny App "ExploreCBS_ShinyApp.R")
6. Jupyter Notebook : Step by step walkthrough of "Explore the CBS Crime and Demographics Dataset.ipynb"

Some screenshots