Skip to content
Snippets Groups Projects

Some changes to the results file for reweighting and uses of reweighted results

Merged Ethan Payne requested to merge (removed):reweighting-changes into master
All threads resolved!

I've made some changes to some of the output saved from reweighting as well as updating some functions to calculate confidence intervals using weights.

Edited by Ethan Payne

Merge request reports

Approval is optional

Merged by avatar (Apr 15, 2025 7:43pm UTC)

Activity

Filter activity
  • Approvals
  • Assignees & reviewers
  • Comments (from bots)
  • Comments (from users)
  • Commits & branches
  • Edits
  • Labels
  • Lock status
  • Mentions
  • Merge request status
  • Tracking
  • Ethan Payne mentioned in merge request !851 (merged)

    mentioned in merge request !851 (merged)

  • Ethan Payne changed title from Some changes to the results file and uses of reweighted results to Some changes to the results file for reweighting and uses of reweighted results

    changed title from Some changes to the results file and uses of reweighted results to Some changes to the results file for reweighting and uses of reweighted results

    • Resolved by Ethan Payne

      Hi @ethan.payne thanks for submitting this. I am bit concerned if this is the best way to implement these changes. You effectively add a bunch of new attributes to the result object that aren't used in 99% of all cases.

      I think it would actually be better either:

      1. reweight returns a new Result object that is different from the one you pass in instead of modifying the one the user provides. You'd have to return the efficiency and weights separately and you would get rid off old/new_log_likelihood/prior_array. These arrays still live in the individual results and can thus be accessed by the user.

      2. You make reweight a method of Result. This would match more naturally with what you are trying to do here. old/new_log_likelihood/prior_array, weights, ... would be attributes that aren't available for the user to set during initialisation then but they would only be set when reweight is being run. Problem with that is that it breaks backwards compatibility with existing code, so I would be cautious implementing that.

  • Moritz Huebner
  • Moritz Huebner
  • Moritz Huebner
  • Moritz Huebner
  • Ethan Payne added 2 commits

    added 2 commits

    • c446c7d3 - changing how the reweighting results are stored/handled
    • f77e00db - flake8

    Compare with previous version

  • Ethan Payne resolved all threads

    resolved all threads

  • Gregory Ashton
  • Gregory Ashton
  • Moritz Huebner changed milestone to %Future

    changed milestone to %Future

  • Moritz Huebner changed milestone to %1.0.4

    changed milestone to %1.0.4

  • Ethan Payne added 1 commit

    added 1 commit

    • 0df7c65c - Updated docstrings and removed conversion function within reweighting code

    Compare with previous version

  • Ethan Payne added 116 commits

    added 116 commits

    • 0df7c65c...d5f739b5 - 114 commits from branch lscsoft:master
    • 64edb2cb - Merge remote-tracking branch 'upstream/master'
    • 00c0ab08 - Merge branch 'master' into reweighting-changes

    Compare with previous version

  • Hey @gregory.ashton, I think the code is almost ready to be merged though it seems to be failing other (unrelated) tests which I am unsure on how to resolve...

  • Colm Talbot mentioned in merge request !917 (merged)

    mentioned in merge request !917 (merged)

  • Ethan Payne added 10 commits

    added 10 commits

    • 00c0ab08...762568df - 9 commits from branch lscsoft:master
    • e10f3a34 - Merge remote-tracking branch 'upstream/master' into reweighting-changes

    Compare with previous version

  • Moritz Huebner resolved all threads

    resolved all threads

  • Moritz Huebner
  • Moritz Huebner mentioned in merge request !918 (closed)

    mentioned in merge request !918 (closed)

  • Moritz Huebner changed milestone to %1.1.0

    changed milestone to %1.1.0

  • Moritz Huebner approved this merge request

    approved this merge request

  • Moritz Huebner resolved all threads

    resolved all threads

  • Matthew Pitkin
  • Ethan Payne added 1 commit

    added 1 commit

    • 50e9d429 - updating result prior dictionary when reweighting

    Compare with previous version

  • Ethan Payne resolved all threads

    resolved all threads

  • Matthew Pitkin approved this merge request

    approved this merge request

  • Ethan Payne resolved all threads

    resolved all threads

  • Ethan Payne added 21 commits

    added 21 commits

    • 50e9d429...bef3cfa5 - 20 commits from branch lscsoft:master
    • 4d63c2c3 - Merge remote-tracking branch 'upstream/master' into reweighting-changes

    Compare with previous version

  • Moritz Huebner approved this merge request

    approved this merge request

  • Shanika Galaudage approved this merge request

    approved this merge request

  • Ethan Payne resolved all threads

    resolved all threads

  • Ethan Payne added 3 commits

    added 3 commits

    • 4d63c2c3...54c90923 - 2 commits from branch lscsoft:master
    • ce23e2b9 - Merge remote-tracking branch 'upstream/master' into reweighting-changes

    Compare with previous version

  • Moritz Huebner approved this merge request

    approved this merge request

  • Shanika Galaudage approved this merge request

    approved this merge request

  • Moritz Huebner mentioned in commit 309158d6

    mentioned in commit 309158d6

  • Moritz Huebner mentioned in merge request !928 (merged)

    mentioned in merge request !928 (merged)

  • Please register or sign in to reply
    Loading