PyData Academy

Initiative of Continuum Analytics (http://continuum.io) to provide knowledge for data scientist curriculum.

Goals:

• Help solve problems in data science
• give insight into data processing for data analysts
• give insight into data analysis for data warehouse engineers

The academy structure is based on modules. Modules are short trainings or presentations and appropriate module content should fit somewhere within 30 minutes to two hours. Finer granularity allows better fitting of overall training for target customer/audience. Each module has:

description of problems the course skills/knowledge will help to solve (note: multiple modules can solve one problem) prerequisites: knowledge (even external, with provided resource links) or other modules follow-up – suggested next modules list of skills that the learner will gain after taking the module exercises and example datasets

Contents

The project contains following folders:

• modules - knowledge/skill modules
• catalogue – module catalogue in presentation form (generated HTML)
• data – shared example datasets and their metadata
• data/datasets.json – metadata about the datasets (see below)

Modules

The module directory contains:

• course description (might be contained in the included Python notebooks)
• python scripts
• one or more python notebooks
• data directory if the datasets are related only to that particular module
• module metadata in module.json

Datasets Metadata

Each dataset is accompained by a file with same with .json suffix. The file contains information about the dataset such as:

• dataset source URL
• date of last file download/update from the source (sometimes required by TOS of the data provider)
• other information, such as list of fields

Reason for datasets metadata are:

• provide links to original dataset source for more information or other similar datasets
• describe the data conteint in more details
• legal requirements

Development

Some PyData academy development notes:

Catalogue

The module catalogue in html form is generated by running following script from project's root directory:

python catalogue/build.py

Page templates are stored in catalogue/templates and files are generated directly into the catalogue directory.

Copyright

(c) 2012 Continuum Analytics.

License: Creative Commons Attribution-ShareAlike 3.0 Unported License

See LICENSE file for more information.