Adjust PairwiseGP ScaleKernel prior #1460
Closed
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
facebook-github-bot
added
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
ItsMrLin
added a commit
to ItsMrLin/botorch
that referenced
this pull request
Oct 20, 2022
Summary: Pull Request resolved: pytorch#1460 Updating the prior of PairwiseGP's output scale prior. Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced `_batch_chol_inv` with `torch.cholesky_inverse`. TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line. Differential Revision: D40136741 fbshipit-source-id: 7ca60923658909b9b2bfd40e4f0456f223ac11f2
ItsMrLin
force-pushed
the
export-D40136741
branch
from
850659b to
1ebbe69
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
Codecov Report
@@ Coverage Diff @@
## main #1460 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 134 134
Lines 12382 12384 +2
=========================================
+ Hits 12382 12384 +2
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
ItsMrLin
added a commit
to ItsMrLin/botorch
that referenced
this pull request
Oct 20, 2022
Summary: Pull Request resolved: pytorch#1460 Updating the prior of PairwiseGP's output scale prior. Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced `_batch_chol_inv` with `torch.cholesky_inverse`. TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line. Differential Revision: D40136741 fbshipit-source-id: 24636bf178d697ed6f2af999b8b7b64551f6283d
ItsMrLin
force-pushed
the
export-D40136741
branch
from
1ebbe69 to
6b1c23d
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
ItsMrLin
added a commit
to ItsMrLin/botorch
that referenced
this pull request
Oct 20, 2022
Summary: Pull Request resolved: pytorch#1460 Updating the prior of PairwiseGP's output scale prior. Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced `_batch_chol_inv` with `torch.cholesky_inverse`. TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line. Differential Revision: D40136741 fbshipit-source-id: c7641d1af33bc89459291234883077005abb1577
ItsMrLin
force-pushed
the
export-D40136741
branch
from
6b1c23d to
52e31be
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
ItsMrLin
added a commit
to ItsMrLin/botorch
that referenced
this pull request
Oct 21, 2022
Summary: Pull Request resolved: pytorch#1460 Updating the prior of PairwiseGP's output scale prior. Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced `_batch_chol_inv` with `torch.cholesky_inverse`. TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line. Reviewed By: Balandat Differential Revision: D40136741 fbshipit-source-id: 7f48b162ef529416f5acbb6045d92b0e28b62255
ItsMrLin
force-pushed
the
export-D40136741
branch
from
52e31be to
6e7a346
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
ItsMrLin
added a commit
to ItsMrLin/botorch
that referenced
this pull request
Oct 21, 2022
Summary: Pull Request resolved: pytorch#1460 Updating the prior of PairwiseGP's output scale prior. Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced `_batch_chol_inv` with `torch.cholesky_inverse`. TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line. Reviewed By: Balandat Differential Revision: D40136741 fbshipit-source-id: 2673441bb7210fee95f3ec981d88d954a794f5a2
ItsMrLin
force-pushed
the
export-D40136741
branch
from
6e7a346 to
871a937
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
Summary: Pull Request resolved: pytorch#1460 Updating the prior of PairwiseGP's output scale prior. Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced `_batch_chol_inv` with `torch.cholesky_inverse`. TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line. Reviewed By: Balandat Differential Revision: D40136741 fbshipit-source-id: 716ff579b4f670ee93703b065f1c3670af2fbbc7
ItsMrLin
force-pushed
the
export-D40136741
branch
from
871a937 to
faa233c
Compare
|
This pull request was exported from Phabricator. Differential Revision: D40136741 |
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.
Summary:
Updating the prior of PairwiseGP's output scale prior.
Additionally, also make sure it must be used, better initialization of the inferred utility values, and replaced
_batch_chol_invwithtorch.cholesky_inverse.TLDR is that we were previously using an significantly restrictive prior on the output scale theta (note theta = 1/sigma^2 where sigma is the probit noise on the function value), this prevent us from accommodating comparison errors outside range of the green line.
Differential Revision: D40136741