Skip to content

Robust computation of the probability weights #12

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

Open
NglQ wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Robust computation of the probability weights #12

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
8 changes: 7 additions & 1 deletion error_parity/classifiers.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -279,11 +279,17 @@ def find_weights_given_two_points(
f"should be 1!"
)

if not all(np.isclose(target_point, all_weights @ all_points)):
if not all(np.isclose(target_point, all_weights @ all_points, atol=1e-5)):
raise RuntimeError(
f"Triangulation of target point failed. "
f"Target was {target_point}; got {all_weights @ all_points}."
)

if (all_weights < 0).any():
Copy link
Member

@AndreFCruz AndreFCruz Sep 10, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Did you ever encounter a scenario where any weights were negative? And if so, could you post a code example?

all_weights = np.asarray(all_weights)
x_max = np.amax(all_weights, axis=0, keepdims=True)
exp_x_shifted = np.exp(all_weights - x_max)
all_weights = exp_x_shifted / np.sum(exp_x_shifted, axis=0, keepdims=True)

return all_weights, all_points

Expand Down