drivendata/data-science-is-software

data-science-is-software's Issues

Move code coverage instructions from collab slides to test slides

Suggestion for alternative subtitle

Amazing stuff, guys!
I'd only suggest a more telling subtitle: instead of Developer #lifehacks for the Jupyter Data Scientist probably something like How you should organize your data science project or How to better organize your data science project

Create lab solutions

Add coverage to testing (and make sure solution works)

Add requirements.txt to this repo with actual requirements

We generally send people to https://drivendata.github.io/pydata-setup/ in order to get their machines running. The .env files linked there don't include some of the dependencies we need for this project (e.g., engarde, tqdm).

Create lab notebook

Add test for underperforming model

Should be pretty easy just using an assert like the second to last cell here:
https://github.com/drivendata/data-science-is-software/blob/master/notebooks/labs/3.0-refactoring-solution.ipynb

Create lab notebook

Docstrings

Explain how to make parse-able doc strings

Environment slides need work

Create lecture notebook

Create lab solutions

Create lab solutions

Create lecture notebook

Port content from old repo

Write README

• link pydata setup instructions
  • direct link to slides
  • TOC

Make repo public

Adjust testing lab to refer specifically to new Refactoring functions

Create lab solutions

Lecture NB 3.0: code profiling

Section 4.2

Create lecture notebook

Create lecture notebook

Save notebook as .py

• Add slide about this
  • bitly the gist

Create lecture notebook

Create lab notebook

Finalize numbering throughout

Create lab solutions

Re-generate "lab" from "solution"

Create lecture notebook

Add dotenv information?

No home for this yet, see section 3.9 here:
https://github.com/pjbull/data-science-is-software/blob/master/notebooks/data-science-is-software-talk.ipynb

Create lab notebook

Lecture NB: signpost python3 for env

Add cookiecutter lecture/lab slides to send of project structure

Great talk, thank you

Loved especially your cookie cutter ds templatate. :)
Suggestion for future recorded talks: repeat questions into the mic.

3.0 Set up pycharm

Create lab solutions

Copy everything to thumb drives

Create lab notebook

ERROR: Double requirement given: jupyter (from -r https://raw.githubusercontent.com/drivendata/data-science-is-software/master/requirements.txt (line 15)) (already in jupyter>=1.0.0 (from -r https://raw.githubusercontent.com/drivendata/data-science-is-software/master/requirements.txt (line 4)), name='jupyter')

The requirements.txt file has duplicate listings. eg: jupyter, ipython.

Verification script for project setup

Add collaboration slides

• Source control
• Git Flow
• Issue tracking with waffle.io
• Autogenerate .py for notebooks
• CI Server + CodeCoverage Saas

Testing: lecture NB

Flake 8 pass

Create lab notebook

Clear all NB output cells