feat(issue summary) Receive optional possible cause scores #84346
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
github-actions
bot
added
the
Scope: Backend
roaga
approved these changes
Jan 30, 2025
kddubey
deleted the
kddubey/issue-summary/thresholds-possible-cause-backend
branch
kddubey
added a commit
that referenced
this pull request
Jan 31, 2025
…84349) ## Background `possible_cause_confidence` increases when there's less speculation (which correlates w/ slightly more accurate causes) `possible_cause_novelty` increases when there's more novelty / less redundancy wrt `whats_wrong` thresholds led to 50% of possible causes getting dropped on our autofix sentry issues ([notebook here](https://github.com/getsentry/data-analysis/blob/main/autofix/issue_summary/issue_summary_confidence.ipynb)) ## Backend changes corresponding backend change to sentry: #84346 corresponding backend change to seer: getsentry/seer#1788 without the sentry backend change (there are no `data.scores`), default to current behavior: always show possible cause  with the change, only show it if both scores are greater than the threshold  (note: this example is actually considered novel and confident-enough. I hardcoded the threshold to test the behavior)
andrewshie-sentry
pushed a commit
that referenced
this pull request
Feb 5, 2025
<!-- Describe your PR here. --> schema from [here](https://github.com/getsentry/seer/blob/696452da11113cec497029bfb92339d7660ead8d/src/seer/automation/summarize/models.py#L17-L28) <!-- Sentry employees and contractors can delete or ignore the following. -->
andrewshie-sentry
pushed a commit
that referenced
this pull request
Feb 5, 2025
…84349) ## Background `possible_cause_confidence` increases when there's less speculation (which correlates w/ slightly more accurate causes) `possible_cause_novelty` increases when there's more novelty / less redundancy wrt `whats_wrong` thresholds led to 50% of possible causes getting dropped on our autofix sentry issues ([notebook here](https://github.com/getsentry/data-analysis/blob/main/autofix/issue_summary/issue_summary_confidence.ipynb)) ## Backend changes corresponding backend change to sentry: #84346 corresponding backend change to seer: getsentry/seer#1788 without the sentry backend change (there are no `data.scores`), default to current behavior: always show possible cause  with the change, only show it if both scores are greater than the threshold  (note: this example is actually considered novel and confident-enough. I hardcoded the threshold to test the behavior)
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
schema from here