Skip to content

Reject a model whose class codes would be scored against label names #9

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
Akram Zaytar (Akramz) merged 1 commit into from
Oct 31, 2026
+5 −0
Merged

Reject a model whose class codes would be scored against label names #9

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions pappi/core/pipeline.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -284,6 +284,11 @@ def _as_model(m, data: Dataset) -> Fitted:
if not hasattr(m, "classes_") and not np.isnan(num).any() and num != sorted(num):
raise ValueError(f"numeric labels sort as strings here {list(data.classes)}; give the "
"model a classes_ attribute listing its column order")
## Codes 0..k-1 stand for the labels only when the labels are numbers: a model fit on
## is_crop 0/1 against labels "crop"/"noncrop" would be scored with its classes swapped.
if names == codes and names != list(data.classes) and np.isnan(num).any():
raise ValueError(f"model classes {names} are codes but the labels are names "
f"{list(data.classes)}; fit the model on the label names")
if names != list(data.classes) and (names != codes or misordered):
raise ValueError(f"model classes {names} != label classes {list(data.classes)}")
cols = getattr(m, "feature_names_in_", None)
Expand Down
Loading