Skip to content

Conversation

@andLaing
Copy link
Collaborator

Only uses positive charge for
weighting to avoid non-sequential
bins.

@andLaing andLaing requested a review from jmbenlloch June 26, 2019 14:03
@andLaing andLaing force-pushed the protect-rebin branch 2 times, most recently from 5c18ea2 to 2696937 Compare July 16, 2019 11:07
@andLaing andLaing force-pushed the protect-rebin branch 2 times, most recently from 9937bad to f63afc8 Compare November 8, 2019 15:07
Copy link
Collaborator

@gonzaponte gonzaponte left a comment

Choose a reason for hiding this comment

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

Perfect! Good way to get rid of the negative weights that were messing up the order of the time bins.

Only uses positive charge for
weighting to avoid non-sequential
bins.
Data had a peak affected by -ve charges
data updated to represent new expectation
@andLaing andLaing force-pushed the protect-rebin branch 2 times, most recently from 420b37c to 1841241 Compare November 26, 2019 14:22
@gonzaponte
Copy link
Collaborator

Are you sure about this last commit? I don't think there is a huge difference in any case, but at first glance, the previous option seems less invasive, meaning that you only do something when it is actually needed.

@andLaing
Copy link
Collaborator Author

Yes, sorry. I realised after the push that it was wrong and undid it. We're seeing some fail paths though due to an issue with the deconv and/or the selection of the peaks. I'mm looking at protections but I think it's a wider issue that should be addressed so I'll put an issue

@andLaing
Copy link
Collaborator Author

andLaing commented Dec 5, 2019

I'm going to close this pull request as we found some failures in high statistics testing.

The changes plus additional protections are revisited in PR #683

@andLaing andLaing closed this Dec 5, 2019
@andLaing andLaing mentioned this pull request Dec 17, 2019
carmenromo added a commit that referenced this pull request Dec 31, 2019
#687

[author: andLaing]

This PR reopens #658 with an added patch to protect against attempts
to rebin using only negative charge bins.

Tests are added to demonstrate the problems and their solution.

[reviewer: mmkekic]

This PR demonstrates and fixes the bug in weighted rebinning of the
waveforms with negative weights. The change is minimal and clear, and
the test are relevant for any kind of future refactoring of irene
city. Good job!
andLaing pushed a commit to andLaing/IC that referenced this pull request May 13, 2020
next-exp#687

[author: andLaing]

This PR reopens next-exp#658 with an added patch to protect against attempts
to rebin using only negative charge bins.

Tests are added to demonstrate the problems and their solution.

[reviewer: mmkekic]

This PR demonstrates and fixes the bug in weighted rebinning of the
waveforms with negative weights. The change is minimal and clear, and
the test are relevant for any kind of future refactoring of irene
city. Good job!
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants