Add integrated histogram visualizations #3942
Conversation
tanujkhattar
requested review from
cduck,
vtomole,
wcourtney and
a team
as code owners
cirq/vis/histogram.py
Outdated
| color: Color of the line. | ||
| linestyle: Linestyle to use for the plot. | ||
| lw: linewidth for integrated histogram | ||
| ms: marker size for a histogram trace |
There was a problem hiding this comment.
Eg:
cirq.integrated_histogram(range(10), marker='o', ms=10.0)
Marker size corresponds to the size of "o"s.
| lw: linewidth for integrated histogram | ||
| ms: marker size for a histogram trace | ||
| label: An optional label which can be used in a legend. | ||
|
|
There was a problem hiding this comment.
hide_zero, title are missing
There was a problem hiding this comment.
added. Also modified hide_zero to show_zero to avoid double negation. (if not hide_zero: do something special)
|
|
||
| if not hide_zero: | ||
| bin_values = np.linspace(0, 1, n + 1) | ||
| parameter_values = sorted(np.concatenate(([0], data))) |
There was a problem hiding this comment.
Can you help me understand this? It looks a bit fishy to me - we are modifying the data to show where zero is? Why not just modify the x axis min value?
There was a problem hiding this comment.
A step plot starts by plotting a horizontal line between the first (x0, y0) and (x1, y0) before going up from (x1, y0) to (x1, y1). Because of this, the first horizontal line would co-inside with x-axis (y values are generated by np.linespace so first y value would always be 0).
If hide_zero is false, we move the plot up by 1 unit. For example, the left plot below is generated using hide_zero = True and right plot is generated using hide_zero = False.
- Left:
cirq.integrated_histogram(range(1, 10), marker='o', ms=10.0, hide_zero=True) - Right:
cirq.integrated_histogram(range(1, 10), marker='o', ms=10.0, hide_zero=False)
tanujkhattar
added
the
automerge
CirqBot
added
the
front_of_queue_automerge
CirqBot
removed
automerge
This PR adds integrated histogram visualizations support to Cirq. Closes #3941
See https://tinyurl.com/cirq-visualizations for the larger roadmap item.
Usage:
calibration.plot('single_qubit_errors')now produces the following image:Next steps
visualizing_calibration_metricstutorial will be updated in a follow up PR once this is checked in.