search

sandialabs / pyGSTi

A python implementation of Gate Set Tomography
http://www.pygsti.info
Apache License 2.0
134 stars 56 forks source link

Per Sequence Detail Bugfix #364

Open pcwysoc opened 11 months ago

pcwysoc commented 11 months ago

Describe the bug The per sequence detail plot in the HTML report should be calibrated such the probability is 5% that the box or boxes corresponding to an individual circuit are red, not that the probability is 5% that any red box appears. This means the transition point from greyscale to rescale should be calculated

image

where N is the number of circuits not the number of boxes. The fix should be in report.workspaceplots.ColorBoxPlot on line 1821 or so.

Environment (please complete the following information):

sserita commented 8 months ago

FYI Erik has a branch about this called feature-percircuit-colorboxplot-transition. We just need to finish out that branch to fix this, but targeting that for the next major release.

pcwysoc commented 8 months ago

Hi Stefan--that branch was a roadmap that Erik sketched out for me when I ran into this issue. My branch has this fix as well as others to handle different numbers degrees of freedom per circuit.

sserita commented 8 months ago

Thanks for the additional info! Then this is my reminder to delete that branch once we get this in.

Also, FYI: I'm imagining getting a bunch of your bugfixes in from the qiskit-mcm branch into develop before the 0.9.13 release (but not for 0.9.12.1). Even if we leave your branch alive for continued development, there is a significant amount of fixes on there that are worth incorporating into develop on the next month or two timescale.