Add support for missing tasks in mtgp #2960
Open
Conversation
This file contains hidden or 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
meta-cla
bot
added
the
CLA Signed
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 8, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
Codecov Report✅ All modified and coverable lines are covered by tests. Additional details and impacted files@@ Coverage Diff @@
## main #2960 +/- ##
=========================================
Coverage 100.00% 100.00%
=========================================
Files 216 216
Lines 20246 20256 +10
=========================================
+ Hits 20246 20256 +10 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 8, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 8, 2025
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
f03dda4 to
fae6912
Compare
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 8, 2025
Summary: X-link: facebook/Ax#4121 Pull Request resolved: pytorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
fae6912 to
3ad186d
Compare
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 8, 2025
Summary: Pull Request resolved: facebook#4121 X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 8, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 8, 2025
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
3ad186d to
5ef3c77
Compare
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 8, 2025
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
5ef3c77 to
023b232
Compare
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 8, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 9, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 9, 2025
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
023b232 to
6c977ee
Compare
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 12, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 12, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 12, 2025
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
6c977ee to
fc0e7fd
Compare
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 12, 2025
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
fc0e7fd to
09dd4dd
Compare
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
1 similar comment
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
sdaulton
added a commit
to sdaulton/botorch
that referenced
this pull request
Aug 12, 2025
Summary: X-link: facebook/Ax#4121 Pull Request resolved: pytorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
09dd4dd to
bd43641
Compare
sdaulton
added a commit
to sdaulton/Ax-1
that referenced
this pull request
Aug 12, 2025
Summary: X-link: pytorch/botorch#2960 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
Summary: X-link: facebook/Ax#4121 Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data. This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair. This will still error out in Ax if data for the target trial is missing. Differential Revision: D79812024
sdaulton
force-pushed
the
export-D79812024
branch
from
bd43641 to
aee2ced
Compare
|
This pull request was exported from Phabricator. Differential Revision: D79812024 |
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:
Currently, cross-validation in Ax fails when using a MTGP if there are multiple metrics and only some metrics have been observed for some tasks. This is a modeling problem, since the model is a ModelListGP and not all MTGPs in the list are required to have the same tasks. Hence when you pass in a test input, the model errors out if there are not observations from that task in the training data.
This avoids the error by mapping (optionally) mapping unexpected tasks to the "target task". This does not change the default behavior. For cross-validation in Ax, predictions are discarded if there are no observations for a given (task, metric) pair.
This will still error out in Ax if data for the target trial is missing.
Differential Revision: D79812024