Knowledge and some other fixes. #1318
Merged
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
…Z and SemBicScore--these were using testwise deletion when they didn't need to. To pull over the SEM BIC code from another branch, I also needed to pull over a modest revision to the FGES code, where I shore up some of the reasoning.
|
Tests pass. The problem with Knowledge was that it wasn't forbidding all necessary backward edges. The problem with Fisher Z and SEM BIC was that if you gave them large datasets they would try to do the slow testwise deletion, even then there were no missing values. The changes for FGES were needed because of the changes to SEM BIC. Basically, a turning step was added to FGES a la PCALG's GES, which is being left off for now. Also, extra checks are added to the preconditions for some of the rules to make sure the rules are applied correctly in every case. Corresponding adjustments were made to Params and the manual. |
…ion editor, where if you click the simulate bug twice you don't get a news simulation. You have to select a new model type. Fixed this. Affected several files. For FGES, decided to add the turning and tdepth parameters. Adjusted the manual.
|
That's it. These bugs fixed:
|
|
Thanks Joe. Will take a look later today.
…On Fri, Jan 8, 2021 at 9:48 AM Joseph Ramsey ***@***.***> wrote:
That's it. These bugs fixed:
- Knowledge.
- Fisher Z and Sem Bic with datasets that don't contain missing values.
- The interface "Simulate" bug.
- Some incidental changes to FGES (the algorithm hasn't changed).
—
You are receiving this because your review was requested.
Reply to this email directly, view it on GitHub
<#1318 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AACMS6OKYMOIHWLN554SN2LSY4LK7ANCNFSM4VZPKRPQ>
.
|
kvb2univpitt
requested review from
kvb2univpitt
and removed request for
espinoj
kvb2univpitt
approved these changes
Jan 25, 2021
|
Thanks! I'll see if Jeremy can post it (as a bug fix).
…On Mon, Jan 25, 2021 at 10:52 AM Kevin Bui ***@***.***> wrote:
***@***.**** approved this pull request.
Tested knowledge, and it seems to be working fine. Ran data with and
without missing data on FisherZ and SemBIC didn't seem to break anything.
—
You are receiving this because you authored the thread.
Reply to this email directly, view it on GitHub
<#1318 (review)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/ACLFSR7MGQZCAFSSY56I5BTS3WHVTANCNFSM4VZPKRPQ>
.
--
Joseph D. Ramsey
Special Faculty and Director of Research Computing
Department of Philosophy
135 Baker Hall
Carnegie Mellon University
Pittsburgh, PA 15213
[email protected]
Office: (412) 268-8063
http://www.andrew.cmu.edu/user/jdramsey
|
|
@espinoj Hey Jeremy--Kevin reviewed this pull request--would you have to post it (as a bug fix branch)? Thanks, Joe |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Mainly fixing a knowledge bug but also fixing issues in IndTestFisherZ and SemBicScore--these were using testwise deletion when they didn't need to. To pull over the SEM BIC code from another branch, I also needed to pull over a modest revision to the FGES code, where I shore up some of the reasoning.