widgets

Q# widgets

The Q# widgets are Jupyter Widgets built using the AnyWidget package https://anywidget.dev/

Build with ./build.py --widgets.

Install the built wheel via

pip install --force-reinstall ./target/wheels/qsharp_widgets-0.0.0-py2.py3-none-any.whl

For development, perform an 'editable' install with pip install -e ./widgets in the venv used for testing. Any changes made to the package are then immediately reflected.

If developing the web code (JS and CSS), then in the test environment also install pip install watchfiles, and in the ./widgets directory run npm run dev to build in watch mode. This will use AnyWidget's hot module reloading to automatically update the Python package as changes are made. (See https://anywidget.dev/blog/anywidget-02/).

With the above done, use the Q# widgets in you Python test environment via import qsharp_widgets.

Usage

In a notebook, generate the estimates for a program and display the widgets with code such as that shown below:

# Cell-1 : Import the modules and generate some estimates
import qsharp
from qsharp_widgets import SpaceChart, EstimateDetails

with open("sample.qs", "r") as f:
    contents = f.read()
qsharp.eval(contents)
result1 = qsharp.estimate("Sample.Main()")

# Cell-2 : Display the details in table form
EstimateDetails(result1)

# Cell-3 : Display the space chart
SpaceChart(result1)

# Cell-4 : Use the logical counts to get estimates for a different qubit
result2 = qsharp.physical_estimates_from_logical_counts(
    result1.get("logicalCounts"),
    {
        "qubitParams": { "name": "qubit_maj_ns_e6" },
        "qecScheme": { "name": "floquet_code" }
    }
)
SpaceChart(result2)