search

githubpsyche / pcitpy

The Probabilistic Curve Induction and Testing (P-CIT) toolbox, implemented in Python.
https://githubpsyche.github.io/pcitpy/
Apache License 2.0
0 stars 0 forks source link

common_to_all_curves testing #4

Closed githubpsyche closed 4 years ago

githubpsyche commented 4 years ago

Must develop and pass a test for each option.

githubpsyche commented 4 years ago

We're given some example input formats, but not example input and output; this means tests have to be imagined ourselves.

githubpsyche commented 4 years ago

initial_sampling fully tested and so is check_if_exceeds_bounds.

Still need curve_volumes, flip_vertical_params, sort_horizontal_params, draw_bcm_curve, auto_generate, weighted_curve - holy shit.

githubpsyche commented 4 years ago

draw_bcm_curve doesn't seem to ever actually get used in the codebase. auto_generate is exclusively used for data simulation. weighted_curve is used for output analysis. The rest get used just once during importance sampling. Maybe I test them as they come?

githubpsyche commented 4 years ago

It'll be easier to work on this if I set up the relevant software locally. What's necessary?

githubpsyche commented 4 years ago

I can use usage within other functions as a guide for testing - I should validate function performance within the contexts they're used.

githubpsyche commented 4 years ago

rebooting discussion in #31