Fix handling of trainable_params in qs.copy #6363
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
andrijapau
approved these changes
Oct 8, 2024
Co-authored-by: Andrija Paurevic <46359773+andrijapau@users.noreply.github.com>
Codecov ReportAll modified and coverable lines are covered by tests ✅
Additional details and impacted files@@ Coverage Diff @@
## master #6363 +/- ##
=======================================
Coverage 99.39% 99.39%
=======================================
Files 446 446
Lines 42326 42328 +2
=======================================
+ Hits 42071 42073 +2
Misses 255 255 ☔ View full report in Codecov by Sentry. |
austingmhuang
pushed a commit
that referenced
this pull request
Oct 23, 2024
**Context:** 1. The existing implementation of `QuantumScript.copy` relies on users to explicitly pass `trainable_params=None` to recalculate trainable params; however, updating `operations` and/or `measurements` often makes the `trainable_params` list outdated. 2. There is a bug that raises an error when explicitly passing `trainable_params=None` to qs.copy **Description of the Change:** 1. If a user passes `trainable_params` explicitly along with `operations` and/or `measurements`, we continue using the user-defined `trainable_params`. However, if updating `operations`/`measurements` and `trainable_params` is not passed, we default to recalculating for the new tape, rather than to copying over the initial tape's `trainable_params` attribute. 2. We stop trying to cast the input `trainable_params` to a list. I think this was intended to make it possible to pass `trainable_params=1`, but in hindsight, this isn't valid input on in `QuantumScript.__init__` and it shouldn't be valid input here, and calling `list(None)` was the source of the error. [sc-75393] --------- Co-authored-by: Andrija Paurevic <46359773+andrijapau@users.noreply.github.com>
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.
Context:
QuantumScript.copyrelies on users to explicitly passtrainable_params=Noneto recalculate trainable params; however, updatingoperationsand/ormeasurementsoften makes thetrainable_paramslist outdated.trainable_params=Noneto qs.copyDescription of the Change:
trainable_paramsexplicitly along withoperationsand/ormeasurements, we continue using the user-definedtrainable_params. However, if updatingoperations/measurementsandtrainable_paramsis not passed, we default to recalculating for the new tape, rather than to copying over the initial tape'strainable_paramsattribute.trainable_paramsto a list. I think this was intended to make it possible to passtrainable_params=1, but in hindsight, this isn't valid input on inQuantumScript.__init__and it shouldn't be valid input here, and callinglist(None)was the source of the error.[sc-75393]