Skip to content

Fix recent ccweights code for uncalculable correlations #2668

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
jbeilstenedmands merged 3 commits into from
May 22, 2024
Merged

Fix recent ccweights code for uncalculable correlations #2668

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
43a12c2
Consistent return type for uncalculable ccs
jbeilstenedmands May 20, 2024
310747b
Add news
jbeilstenedmands May 21, 2024
76bcfba
Rename newsfragments/XXX.bugfix to newsfragments/2668.bugfix
DiamondLightSource-build-server May 21, 2024
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
1 change: 1 addition & 0 deletions newsfragments/2668.bugfix
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1 @@
``dials.cosym``: Make function return structure correct in recently added cc_weights option
6 changes: 3 additions & 3 deletions src/dials/algorithms/scaling/scaling_library.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -494,7 +494,7 @@ def weighted_cchalf(
if not use_binning:
assert other.indices().size() == this.indices().size()
if this.data().size() == 0:
return None, None
return [(None, 0)]

if assume_index_matching:
(o, c) = (this, other)
Expand All @@ -507,7 +507,7 @@ def weighted_cchalf(
assert len(c.sigmas())
n = len(o.data())
if n == 1:
return None, 1
return [(None, 1)]
v_o = flex.pow2(o.sigmas())
v_c = flex.pow2(c.sigmas())
joint_w = 1.0 / (v_o + v_c)
Expand All @@ -522,7 +522,7 @@ def weighted_cchalf(
sx = flex.sum(flex.pow2(dx) * norm_jw)
sy = flex.sum(flex.pow2(dy) * norm_jw)
if sx == 0.0 or sy == 0.0:
return None, 0
return [(None, 1)]
# effective sample size of weighted sample
# Kish, Leslie. 1965. Survey Sampling New York: Wiley. (R documentation)
# neff = sum(w)^2 / sum(w^2). But sum(w) == 1 as normalised already
Expand Down