https://github.com/tidepool-org/data-science-handbook
Tidepool's Data Science Team Handbook
https://github.com/tidepool-org/data-science-handbook
Last synced: 23 days ago
JSON representation
Tidepool's Data Science Team Handbook
- Host: GitHub
- URL: https://github.com/tidepool-org/data-science-handbook
- Owner: tidepool-org
- License: bsd-2-clause
- Created: 2020-05-01T13:45:44.000Z (almost 5 years ago)
- Default Branch: master
- Last Pushed: 2020-12-16T13:42:08.000Z (over 4 years ago)
- Last Synced: 2025-02-08T22:26:24.526Z (3 months ago)
- Size: 153 KB
- Stars: 1
- Watchers: 7
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Tidepool's Data Science Team Handbook
Welcome to the Tidepool Data Science Team Handbook. Please be sure to read [Team Governance](governance.md)
to get a sense of our purpose and philosophies.
## Purpose
The purpose of this repository is provide guidance to those contributing to the Tidepool Data Science Team,
and to document how Tidepool's Data Science Team operates. It is also a place to capture all of the non-code related
items that make up our ecosystem.
## In this Repository
* [Governance](./GOVERNANCE.md)
* [Best Practices](./best-practices.md)
* [Cheat Sheets](cheat-sheets)
* [Checklists & Templates (empty)](checklists-and-templates)
* G-Suite
* G-Suite Instructions (TODO: instructions/g-suite.md)
* [G-Suite Acrchitecture](g-suite-architecture)
* GitHub
* [GitHub Instructions](instructions/github.md)
* [GitHub Architecture](github-architecture)
* [Instructions](instructions)
* [Tips & Tricks](tips-and-tricks)
* [Tutorials](tutorials)
## Architecture, Ecosystem, & Tech Stack
* 1password - for password management
* aws - for cloud computing
* compute-1 - the original compute-1 was spun down and replaced with the old compute-2
* S3
* confluence - our knowledge-management-system for internal and external knowledge sharing
but it will point to github and g-suite
* datasets
* can contain phi, or psi (partner sensitive information)
* can exist in multiple locations
* can exist in multiple states (e.g., raw, interim, processed)
* for tidepool it can be all data or donor_data
* mvp is to create for internal, partner, and public use
* needs version control
* source can be tidepool or external
* we need to know what transformations were done
* raw data is in MongoDB Atlas
* several processed datasets in S3
* delivered datasets to partners currently sit in google drive
* email - for corresponding with partners and non-time sensitive materials
* g-suite - for creating documents that need review or collaboration
* github - for version control
* jira - for project and sprint management)
* knowledge-management-system - we may use confluence to share
but data will be version controlled in github and g-suite:
* data-science-insights repository (in github)
* Data Science Team Drive (in g-suite)
* local-machine - the tools most of the team uses
* macbook pro
* excel - for viewing tabular
* gitkraken - has a not gui for git
* pycharm - ide with lots of nice features
* meetings - zoom for video chats, and google docs for meeting notes
* monday - for long term and roadmap planning
* slack - for real-time communication
* travis - for continuous integration
* templates and checklists:
* if code or text based put in github,
* if application specific put directions and locations in handbook templates and checklist section
* zoom - for meetings
* zotero - for collecting papers and citing our work
## FAQs
### Google Doc vs. Report in GitHub
Q: When writing a document or report should I use a Google Doc or put the Github via markdown?
A: If the document is going to require feedback (comments, edits, suggestions) then start with a google doc.
Once the document is finalized, it can be converted into a markdown document and stored in GitHub using this
[Google Docs add-on tool](https://gsuite.google.com/marketplace/app/docs_to_markdown/700168918607).