Enhancement/issue 822 #825
Conversation
|
Thanks a lot @aGuyLearning, awesome! Comments as per review :) |
| model = model_class() | ||
| model.fit(T, event_observed=E) | ||
| model.fit( |
There was a problem hiding this comment.
Add fit_left_censoring call upon argument input left
There was a problem hiding this comment.
check if nelson-aalen gracefully crashes with meaningful error message
| durations: Iterable, | ||
| event_observed: Iterable | None = None, | ||
| timeline: Iterable = None, | ||
| entry: Iterable | None = None, |
There was a problem hiding this comment.
no kwargs, but full arguments listed :)
| @@ -376,7 +385,9 @@ def log_logistic_aft(adata: AnnData, duration_col: str, event_col: str, entry_co | |||
| ) | |||
|
|
|||
|
|
|||
| def _univariate_model(adata: AnnData, duration_col: str, event_col: str, model_class, accept_zero_duration=True): | |||
| def _univariate_model( | |||
| adata: AnnData, duration_col: str, event_col: str, model_class, accept_zero_duration=True, **kwargs | |||
There was a problem hiding this comment.
not kwargs but pass arguments, see above
| alpha: float | None = None, | ||
| ci_labels: tuple[str, str] = None, | ||
| weights: Iterable | None = None, | ||
| censoring: Literal["right", "left"] = None, |
There was a problem hiding this comment.
add comment that and "right" just calls fit and "left" calls fit_left_censoring, maybe change default value to "right" instead of None
| entry: Iterable | None = None, | ||
| label: str | None = None, | ||
| alpha: float | None = None, | ||
| ci_labels: tuple[str, str] = None, |
There was a problem hiding this comment.
Add None to typehint to make clear its optional
| @@ -116,15 +116,10 @@ def glm( | |||
|
|
|||
|
|
|||
| def kmf( | |||
There was a problem hiding this comment.
one additional thing here; we'll also have to watch out for backwards compatibility.
Here, we could while still accepting the arguments raise a deprecation warning, where we briefly highlight the new behaviour.
Description of changes
closes #822
Changed the method signature of univariate models to allow the user to pass parameters to the lifeline. Now weilbull, neil_aalen and kmf look like
adata: AnnData, duration_col: str, event_col: str, **kwargsand are passed on to the _univariate_model function
Technical details
this could be changed to explicitly have all parameters in each function and pass them on.
Additional context