-
-
Notifications
You must be signed in to change notification settings - Fork 24
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
Pareto-smoothing updates #314
Conversation
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 7ea8037 is merged into master:
|
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.
PR looks good to me. I just have one request for changes.
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 7a109a6 is merged into master:
|
This is how benchmark results would change (along with a 95% confidence interval in relative change) if ec95ae6 is merged into master:
|
I would prefer as_one_logical as I also use for the log argument.
github-actions[bot] ***@***.***> schrieb am Mi., 15. Nov.
2023, 17:50:
… This is how benchmark results would change (along with a 95% confidence
interval in relative change) if ec95ae6
<ec95ae6>
is merged into master:
- ☑️as_draws_array: 99.6ms -> 100ms [-0.67%, +2.23%]
- ☑️as_draws_df: 31.6ms -> 31.8ms [-0.66%, +1.88%]
- ☑️as_draws_list: 183ms -> 182ms [-1.3%, +1.06%]
- ☑️as_draws_matrix: 29.2ms -> 28.9ms [-2.54%, +0.74%]
- ☑️as_draws_rvars: 155ms -> 157ms [-1.5%, +3.85%]
- ☑️summarise_draws_100_variables: 712ms -> 711ms [-0.54%, +0.16%]
- ☑️summarise_draws_10_variables: 79.3ms -> 79ms [-1.29%, +0.33%]
Further explanation regarding interpretation and methodology can be
found in the documentation
<https://lorenzwalthert.github.io/touchstone/articles/inference.html>.
—
Reply to this email directly, view it on GitHub
<#314 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ADCW2AF4A34WX2FTO64R7TTYETXEHAVCNFSM6AAAAAA7MLANH2VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTQMJSHEYDANBSGQ>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Ready to merge from my side. Anything you want to change before I merge? |
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 2cebd0d is merged into master:
|
I think I should add some tests for the new functionality. Also, I am now just adding a message if khat too high when adding weights. So a couple little things to add. I'll write here when done |
Codecov ReportAttention:
Additional details and impacted files@@ Coverage Diff @@
## master #314 +/- ##
==========================================
- Coverage 95.52% 95.28% -0.25%
==========================================
Files 47 47
Lines 3691 3734 +43
==========================================
+ Hits 3526 3558 +32
- Misses 165 176 +11 ☔ View full report in Codecov by Sentry. |
This is how benchmark results would change (along with a 95% confidence interval in relative change) if f3a56a9 is merged into master:
|
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 2018039 is merged into master:
|
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 0081eab is merged into master:
|
I've now added tests and a message if pareto-k is high. I think it's ready from my side |
This is how benchmark results would change (along with a 95% confidence interval in relative change) if 7ff4d6b is merged into master:
|
Perfect. Looks good. Merging now! |
Summary
pareto_smooth
and related functions to operate on log weights (Pareto smoothing log weights #313)weight_draws
to Pareto-smooth weights (Pareto smoothing log weights #313)r_eff
default to 1, rather thanNULL
Discussion points
What should we do with khat when using
weight_draws
? Should we add a message or warning if khat > 0.7 (or khat_threshold)?TODO
Copyright and Licensing
By submitting this pull request, the copyright holder is agreeing to
license the submitted work under the following licenses: