-
Notifications
You must be signed in to change notification settings - Fork 26
Allow RelionSource to be double precision #1079
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.
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
Conversation
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## develop #1079 +/- ##
========================================
Coverage 88.82% 88.82%
========================================
Files 126 126
Lines 12024 12028 +4
========================================
+ Hits 10680 10684 +4
Misses 1344 1344 ☔ View full report in Codecov by Sentry. |
|
Closes #871 . |
|
The 380px reconstruction of 10073 just completed successfully with doubles using a similar branch. I will setup to retest with the exact code in this PR. (Experiment was started before the code/PR were made). |
j-c-c
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good! It's nice it only took a couple lines. Just one question.
j-c-c
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM!
janden
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks great. Thanks!

The
MeanEstimationfails to converge at higher resolutions for experimental data (189 and higher) in single precision. I believe this is probably related to the accuracy of the basis in these conditions, but there is more work to be done there at another time.This small PR allows
RelionSourceto upcast images and supporting data as doubles to be consumed by other components. It adds some tests checking the dtype override occurs.This resolved for 189 pixel reconstructions and is currently being tested at 380. Will update, WIP draft for now.