From this directory:
- cd cpp
- make all
- cd ../
- python plot_orbitals.py
Step 1: Add unit test checking that python was correctly getting points from the c++ code 1a: Determine point generation method 1b: Write c++ code for point evaluation
Step 2: Make a few example python plots with MatPlotLib to practice & demonstrate plotting techniques with matplotlib
Step 3: Refactor previous class code to my purposes and preferences - Side note: personally I'm pretty big on no abbreviated variable names when I'm authoring something I lose confidence in this frequently haha
Step 4: Iterate through many different plotting techniques Top 2 choices currently: marching_cubes algorithm scatter (but panning has a bug)
Step 5: Plot Atomic orbitals 5a: Test on H2 5b: Refactor, test on CH
Step 6: Plot Molecular orbitals 5a: Decide visual representation of scalar MO coefficients
Molecule/Atom symbols will be capitalized: Class names will be capitalized, pretty much everything else will be lower case and underscore separated.
Marching Cubes Algorithm for Isosurface plotting: https://scikit-image.org/docs/stable/auto_examples/edges/plot_marching_cubes.html
Open bug exists in matplotlib 3d scatter plot opacity: https://stackoverflow.com/questions/71904575/matplotlib-3d-scatter-plot-alpha-varies-when-viewing-different-angles references matplotlib/matplotlib#22861
Challenging to keep grid transformations straight between c++ and python grids/arma::cubes/np arrays, linspaces for subsequent use in matplotlib, etc
To compile the code, enter the command make all under the cpp_src directory.
plot_orbitals.py should provide a rudimentary interface for calling that executable as a subprocess & plotting that data