numpy / user-survey-2020-details

Analysis and publication of NumPy community survey results
https://numpy.org/user-survey-2020-details/
Other
2 stars 6 forks source link
community development numpy survey

NumPy Community Survey Results

Analyses and publications of the results of the NumPy community surveys NumPy Community Survey.

Running an analysis

The following steps are required to analyze and publish the survey results:

  1. Acquiring the survey data
  2. Configuring the environment
  3. Running the analysis & generating the website

Each of these steps is detailed below.

Survey data

Coming soon...

Environment setup

The necessary packages for analyzing and publishing the results are listed in requirements.txt and site/requirements.txt, respectively. The following procedure can be used to set up an environment for data analysis:

python -m venv survey_analysis
source survey_analysis/bin/activate
python -m pip install .[publish]

Running the analysis

Sphinx is used to automatically run the analysis via the myst-nb plugin and generate a static website presenting the results:

cd site
make html

The generated site can then be viewed by opening site/_build/html/index.html in your preferred browser.

Converting to Jupyter notebooks

For those who are more comfortable with Jupyter notebooks, the myst-nb text-based notebook files can be converted to .ipynb files:

cd site
make notebooks

This converts all of the .md files in content/ to .ipynb files of the same name and moves them to site/notebooks. The notebooks can then be opened for interactive use in the standard way, e.g.

cd site/notebooks
jupyter notebook

Note:

The text-based notebook format includes more features than are currently supported by traditional Jupyter notebooks. Therefore, some features (e.g. glue or other MyST roles & directives) will not render when interacting with the .ipynb file.