Workshop on code compatible with Reproducible Research using Jupyter notebooks
- https://github.com/UBEC/ReproducibleResearch/
- https://reproducible-science-curriculum.github.io/rr-jupyter-workshop/
- https://github.com/sara-nl/jupyter-bigdata-notebooks
-
Get code ready in your environment
-
Go through Data carpentry intro
introduction-RR-Jupyter/notebooks/Intro-to-reproducible-research.ipynb- introduction-RR-Jupyter/notebooks/Jupyter_Intro_Background.ipynb
-
A versioned jupyter notebook
- fork https://github.com/UBEC/ReproducibleResearch
- clone your own version of ReproducibleResearch
- open VersionedNotebook through jupyter interface
- edit the code to perform the desired analysis on the Human IDs
- commit the code to the repository with a tag
- run analysis again with different settings for Mouse
- [origin=False]
- commit the code to the repository with a different tag
- check out the versions on github
- https://github.com/UBEC/ReproducibleResearch/compare/v2.0...v3.1
GIT code example
! git commit -a -m "ID version"
! git tag -a v1.4 -m "ID version"
! git push origin v1.4
-
bio-break -
See some SPARK in action
- jupyter-bigdata-notebooks/notebooks/02-spark-intro.ipynb
- jupyter-bigdata-notebooks/notebooks/03-spark-dataframes.ipynb
Use Git aware text-editors for libraries/readmes etc
Do not put everything under git control:
echo ".ipynb_checkpoints" >> .gitignore
- https://gist.github.com/pbugnion/ea2797393033b54674af
- http://stackoverflow.com/questions/18734739/using-ipython-notebooks-under-version-control
Plotting, embedding, etc...
- https://blog.dominodatalab.com/lesser-known-ways-of-using-notebooks/
- https://plot.ly/python/ipython-notebook-tutorial/