Drag analysis #732
Merged
Drag analysis #732
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
wshanks
requested changes
Mar 10, 2022
There was a problem hiding this comment.
Can you add some more test cases? In particular, I think it would be good to add:
- A case where the amplitude varies only between 0 and 0.5 to account for the SX calibration
- A case where the ideal beta is more than half a period away from 0 for the 1 rep case.
Regarding the second case, here is an example that fits well with the current code but not well with this PR:
from qiskit_experiments.library import RoughDrag
from qiskit_experiments.test.mock_iq_backend import DragBackend
from qiskit import pulse
from qiskit.circuit import Parameter
from qiskit.pulse import DriveChannel, Drag
import numpy as np
beta = Parameter("β")
with pulse.build(name="xp") as xp:
pulse.play(Drag(duration=160, amp=0.208519, sigma=40, beta=beta), DriveChannel(0))
backend = DragBackend(gate_name="Drag(xp)", ideal_beta=9)
drag = RoughDrag(1, xp, betas=np.linspace(-10, 10, 101))
expdata = drag.run(backend)I believe this is because this PR allows for both signs of the amplitude which leads to the fit trying to match the peak in the 1 rep case instead of the dip. -- This is actually because the range of beta values is restricted to less than the swept range in the PR.
One other minor comment -- the DragBackend is pretty misleading. When checking for amp sign fitting issues, it took me a while to figure out what the scaling factor is on error. It gets scaled bysum(circuit.count_ops().values()) which is 2 x drag + 2 x u1 + barrier + measure = 6. It feels like the error should get scaled by the number of drag gates instead, if not drag gates / 2.
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
…xperiments into drag_analysis_fit_guess
wshanks
reviewed
Mar 11, 2022
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
…ysis.py Co-authored-by: Will Shanks <[email protected]>
…ysis.py Co-authored-by: Will Shanks <[email protected]>
…ysis.py Co-authored-by: Will Shanks <[email protected]>
Co-authored-by: Will Shanks <[email protected]>
* Made a nasty test less nasty.
wshanks
requested changes
Mar 15, 2022
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
…ysis.py Co-authored-by: Will Shanks <[email protected]>
…ysis.py Co-authored-by: Will Shanks <[email protected]>
wshanks
approved these changes
Mar 16, 2022
qiskit_experiments/library/characterization/analysis/drag_analysis.py
Outdated
Show resolved
Hide resolved
…ysis.py Co-authored-by: Will Shanks <[email protected]>
7 tasks
chriseclectic
added
the
Changelog: Bugfix
paco-ri
pushed a commit
to paco-ri/qiskit-experiments
that referenced
this pull request
Jul 11, 2022
* The DragAnalysis class has been made more robust. The initial guess for the fit parameters have been improved and more tests were added. Furthermore, a fit post processing hook has been added to curve analysis to allow Drag to report drag parameters in the `+-1 / freq` range. Co-authored-by: Will Shanks <[email protected]>
Closed
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
This PR makes the Drag analysis less prone to fit failures by improving the initial guesses.
Details and comments
The initial guesses for the fit amplitudes are adjusted as well as for the base of the fit function. This addresses #685 . This PR also changes the internals of the drag analysis: the three frequencies are replaced with a single frequency which is multiplied by the number of drag gate repetitions. This is done by introducing
repsas a fixed parameter. Furthermore, the frequency is estimated based on the fastest oscillation. Furthermore this PR introduces several tests of challenging data: