Demo01: Data access and reproducible packages (Hutchinson)
Overview
Teaching: 30 min
Exercises: 0 minQuestions
How can you access the data and analyze them? How can you make your analysis reproducible?
Objectives
Explain how fedreated login system work and CADRE data policy.
Explain how CADRE GUI-query builder can help find the data you want.
Explain how one can explore and analyze query result in notebooks.
Demonstrate the CADRE marketplace.
Explain how tools can be built from notebook code
Explain how archives can be created from query results
Explain how to combine tools with archives into reproducible packages
Getting Started
- Login with your BTAA institutional account if you wish to use Web of Science data
- Start your Jupyter Notebook from https://cadre.iu.edu/gateway/jupyter
Notebooks and script files
Notebooks and source code will be automatically downloaded to your personal Jupyter Notebook. They can also be found at the GitHub repository.
The notebook illustrates a simple analysis step by step.
Notebook: AlphaDemoCode/Demo 01/Feb_demo_01.ipynb
The following python script contains the same content but is modified for building CADRE tools.
Script: AlphaDemoCode/Demo 01/Feb_demo_02.py
A requirement file is one way to include additional python packages
AlphaDemoCode/Demo 01/requirements.txt
CADRE packages and tools
All packages, tools and archives in this demo will be created by you. They can be created and accessed from the CADRE Marketplace. https://cadre.iu.edu/gateway/rac
Key Points
Web of Science and Microsoft Academic Graph have different output fields
Use the preview functionality in the GUI-query builder to quickly explore the data
All your query and package results can be viewd in your notebook environment
Naming your queries, tools, archives and packages will make keeping track of them much easier
Mix and match tools with archives to explore combinations of code and data