Conversation
tupui
left a comment
tupui
left a comment
There was a problem hiding this comment.
Thanks for working on this @justushelo !
Could you add a/some test which would have been failing without these changes and now pass?
| inputs : DataFrame of shape (n_runs, n_factors) | ||
| Input variables. | ||
| output : DataFrame of shape (n_runs, 1) | ||
| output : DataFrame of shape (n_runs, 1) or (n_runs,) |
There was a problem hiding this comment.
Why doing the change? I personally find it easier to have a consistent shape.
There was a problem hiding this comment.
Yeah I forgot to ask about that one. I added .flatten() to the parameter output = output.flatten() for both sensitivity_indices.py and decomposition.py as I think it can increase its robustness. Now when the function is called the output parameter can have both output = df['target'] and output = df[['target']]. Just forgot to ask that should it be changed in the docstring to include the "or (n_runs,)". Or if you find this change to do any good.
| foe : ndarray of shape (n_factors, 1) | ||
| First-order effects (also called 'main' or 'individual'). | ||
| soe : ndarray of shape (n_factors, 1) | ||
| soe_full : ndarray of shape (n_factors, 1) |
There was a problem hiding this comment.
I thought it could help any future debugging to have the same variable name soe_full as it can be found in the Matlab function.
2 participants
Fixed sensitivity indices SOE calculation and additionally fix decomposition logic to follow Matlab.