Working files for 2015 edition of the Berkeley Practical Neuroimaging course

See http://practical-neuroimaging.github.com

This repository has files for the exercises during the class.

See the LICENSE file in each directory for copyright and licensing.

Making exercise notebooks from solution notebooks

We usually write the exercise notebooks as solutions first, and then use an automated tool to strip out the solutions and generate the exercise notebook.

This involves some kludgy markup to the code comments starting each code cell.

See:

• day12/smoothing_solutions.ipynb
• day12/smoothing.ipynb

for an example. I generated day12/smoothing.ipynb automatically from day12/smoothing_solutions.ipynb with:

cd day12
../tools/unsolve_nb.py smoothing_solutions.ipynb smoothing.ipynb

unsolve_nb.py always leaves comments at the top of a code call, and will also remove any code from a code cell unless the first comment in the cell starts with # -. So, unsolve_nb.py will leave the top comments but remove all the code in a cell like this:

# fwhm2sigma
# Maybe some some comments here
def fwhm2sigma(fwhm):
    return fwhm / np.sqrt(8 * np.log(2))

It will leave all the cell (code and comments) in a cell like this:

# - get middle line from middle slice
mid_line = mid_slice[:, 25]
plt.plot(mid_line)

So, the procedure for writing an exercise notebook is usually:

• write some_exercises_solutions.ipynb, putting # - comments for cells where you want the students to have the comments and code, and # comments at the top of other cells to give students a clue how to do the exercise;
• when done, save some_exercises_solutions.ipynb and run ../tools/unsolve_nb.py some_exercises_solutions.ipynb some_exercises.ipynb;
• read through some_exercises.ipynb to check the result is what you expect;
• commit both some_exercises_solutions.ipynb and some_exercises.ipynb to the repository;
• make all edits to some_exercises_solutions.ipynb and regenerate some_exercises.ipynb for your edits.