larch.model.troubleshooting.nothing_chosen_but_nonzero_weight#

nothing_chosen_but_nonzero_weight(model, repair: Literal['?', '-', '*', '!'] = '?', verbose=3)[source]#

Check if some observations have no choice but have some weight.

Parameters:

  • model (BaseModel) – The model to check.

  • repair ({'?', '-', '*', '!'}) – How to repair the data. Minus (‘-’) will make the weight zero when there is no choice. Star (‘*’) will also make the weight zero, plus autoscale all remaining weights so the total of the case weights equals the number of cases. A question mark (‘?’) effects no repair, and simply emits a warning.

  • verbose (int, default 3) – The number of example rows to list for each problem.

Returns:

  • model (BaseModel) – The revised dataframe

  • diagnosis (pd.DataFrame) – The number of bad instances, by alternative, and some example rows.

Raises:

ValueError – If the repair is set to ‘!’ and there are any conflicts found.