You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
While trying to write documentation for this module, these are some of the implementation errors I ran into:
the logdet function, as currently implemented according to this source, requires a features x mols matrix, while elsewhere in the package our standard is a mols x features matrix. Do I change the implementation or just note this in the docs?
shannon_entropy() is not correctly implemented, and does not match the cited paper/equation. There is a comment to add a raise error, but reading through the cited paper I see no reason why there would be a value error in its calculation. Do you want me to rewrite this function?
total_diversity_volume is also not implemented according to its paper. The function currently returns the degree of overlap between molecule neighborhoods, not the total diversity volume. I think we should rename the function instead of altering the implementation, as in the paper, total diversity volume does not describe a sub-set of data points, but the entire input space.
The Wasserstien distance function wdud() needs to be rewritten
See: #121
For each function (mentioned in #121), make sure that:
The text was updated successfully, but these errors were encountered: