Revamp handling of base_samples - #1
Closed
Balandat wants to merge 1 commit into
Closed
Conversation
Summary: - revamps usage of base_samples, so that we don't have to do the weird thing of passing tuples of tensors and ints around - simplifies batch_mode_transform - takes some steps in supporting arbitrary batch shapes (though not all) - cleans up documentation Reviewed By: sdaulton Differential Revision: D14280371 fbshipit-source-id: f9dc7ebd5b1438ce508d8677b732082596bb7ee6
facebook-github-bot
added
the
CLA Signed
Contributor
Author
|
This was supposed to be exported to botorch, not AE. |
facebook-github-bot
pushed a commit
to meta-pytorch/botorch
that referenced
this pull request
Mar 4, 2019
Summary: Pull Request resolved: #12 Pull Request resolved: facebook/Ax#1 - revamps usage of base_samples, so that we don't have to do the weird thing of passing tuples of tensors and ints around - simplifies batch_mode_transform - takes some steps in supporting arbitrary batch shapes (though not all) - cleans up documentation Reviewed By: sdaulton Differential Revision: D14280371 fbshipit-source-id: 470d637975b6e27e14c14f94133f41dcffd742b8
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 4, 2019
Summary: Pull Request resolved: meta-pytorch/botorch#12 Pull Request resolved: #1 - revamps usage of base_samples, so that we don't have to do the weird thing of passing tuples of tensors and ints around - simplifies batch_mode_transform - takes some steps in supporting arbitrary batch shapes (though not all) - cleans up documentation Reviewed By: sdaulton Differential Revision: D14280371 fbshipit-source-id: 470d637975b6e27e14c14f94133f41dcffd742b8
Balandat
added a commit
to meta-pytorch/botorch
that referenced
this pull request
Apr 26, 2019
Summary: Pull Request resolved: #12 Pull Request resolved: facebook/Ax#1 - revamps usage of base_samples, so that we don't have to do the weird thing of passing tuples of tensors and ints around - simplifies batch_mode_transform - takes some steps in supporting arbitrary batch shapes (though not all) - cleans up documentation Reviewed By: sdaulton Differential Revision: D14280371 fbshipit-source-id: 470d637975b6e27e14c14f94133f41dcffd742b8
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 22, 2021
Summary: Currently, it’s impossible to create an experiment with both regular Metrics/Data and MapMetrics/MapData. If you try, you’ll run into an error. This series of diffs aims to not only address this problem but also to improve general developer ergonomics through the creation of a new MapData class (MilesMapData). Over the next few diffs we will: - [x] Create MilesMapData - [ ] Replace usages of MapData with MilesMapData - [ ] Reap MapData - [ ] Rename MilesMapData to MapData This diff specifically has created the MilesMapData class (which has nearly identical user facing function signatures to MapData for easy integration) and some associated tests. The general ideas behind MilesMapData are as follows: 1. Data :> MilesMapData. 2. Data will be downcastable to in an ergonomic way, most often f: Data, MilesMapData -> MilesMapData (the combination of this and #1 will solve the mixed data experiment problem). 3. MilesMapData will be able to materialize a dataframe of the same shape as Data (this will allow us to use MilesMapData where Data is used just via polymorphism, without any messy glue code). The flow for creating a so-called mixed data experiment would go somthing like this: A user makes an Experiment with a MilesMapData object. They fill the experiment with mixed data via MilesMapData.from_multiple_data(left, right) and regular data will have its missing map fields inferred automatically. The user does whatever analysis they want with the MilesMapData (early stopping, etc.). Finally, the user uses the MilesMapData as Data via MilesMapData.df (a property method that materializes a Data shaped dataframe) however they want (extract best points, plots, etc.). NOTE: Please cmd-F "TODO" to see some specific inline design questions I would like feedback on. Reviewed By: bernardbeckerman Differential Revision: D32004210 fbshipit-source-id: f426c5d09b07943fb3920137224b5aa6a76db58a
ItsMrLin
added a commit
to ItsMrLin/Ax
that referenced
this pull request
Jan 21, 2025
Summary: We pass kwargs to ModelSpec.fit in order to pass state extracted when this ModelSpec was used last, back into the ModelSpec, in GenNode.fit (https://fburl.com/code/uic21yuw). Along with these, we pass any kwargs manually specified to GenNode.fit, but this behavior is unused in GenStrategy (https://fburl.com/code/20kpa363) and can thus be reaped. That was a remnant of how we used to pass the state from facebook#1, in GenStrategy and not in GenNode (https://www.internalfb.com/diff/D32003890?dst_version_fbid=499298561323086&transaction_fbid=623265442134398) and should be deprecated. Created from CodeHub with https://fburl.com/edit-in-codehub Reviewed By: saitcakmak Differential Revision: D67803329
1 task
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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
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:
Reviewed By: sdaulton
Differential Revision: D14280371