Skip to content

Operation

The package ships with a CLI called acli (autocalibration command line interface) to solve some common tasks that appear quite often. In the following there are a number of useful commands, but if you want to find out information about commands in your shell use acli --help.

Since some of the commands below are using autocompleting in most cases you would have to enable that feature in your shell by running:

1
acli --install-completion

Usually the shell would just suggest you file paths when pressing the tabulator, but with autocompletion enabled, you would be even get suggestions for node names or other inputs depending on your configuration.

CLI

This section provides an overview of the Command Line Interface (CLI) options and their functionalities.

Main Commands

The autocalibration CLI is organized into several main command groups:

  • start: Run the automatic calibration according to the provided config files
  • cluster: Handle operations related to the cluster
  • spi: Handle operations related to the spi rack
  • node: Handle operations related to the node
  • graph: Handle operations related to the calibration graph
  • config: Load and save the configuration files
  • browser: Will open the dataset browser, which makes you view the datasets from measurements
  • redis: Operations to handle redis values
  • joke: Handle operations related to the well-being of the user

Calibration Commands

start

Starts the calibration supervisor.

Usage:

1
acli start [OPTIONS]

Options:

  • -d TEXT: Dummy mode. The calibration chain runs as normal but the returned datasets are dummy. The generation of each dummy dataset takes into account the existing redis config and the provided samplespaces.
  • -c TEXT: Cluster IP address (if not set, it will use CLUSTER_IP from the .env file)
  • -r TEXT: Rerun an analysis (specify the path to the dataset folder)
  • -n, --name TEXT: Specify the node type to rerun (works only with -r option)
  • --browser: Will open the dataset browser in the background and plot the measurement results live

Cluster Commands

cluster reboot

Reboots the cluster.

Usage:

1
acli cluster reboot

This command will prompt for confirmation before rebooting the cluster, as it can interrupt ongoing measurements.

SPI Rack Commands

spi status

Prints information about the current currents on the spi rack. Only couplers that are having a DAC in the SPI configuration are considered. Please check the documentation about configuration about how to change the SPI DACs.

Usage:

1
acli spi status

spi status

Resets all currents on the SPI rack to 0. Only couplers that are having a DAC in the SPI configuration are considered. Please check the documentation about configuration about how to change the SPI DACs.

Usage:

1
acli spi reset

Node Commands

node reset

Resets all parameters in Redis for the specified node(s).

Usage:

1
acli node reset [OPTIONS]

Options:

  • -n, --name TEXT: Name of the node to be reset in Redis (e.g., resonator_spectroscopy)
  • -a, --all: Reset all nodes
  • -f, --from_node TEXT: Reset all nodes from the specified node in the chain

Graph Commands

graph plot

Plots the calibration graph to the user-specified target node in topological order.

Usage:

1
acli graph plot

This command visualizes the calibration graph using an arrow chart.

Configuration Commands

config load

Load the configuration.

Usage:

1
acli config load [OPTIONS]

Options:

  • -f/--filepath: Path to the configuration package to load. It can be either to the configuration.meta.toml file or to a zip file containing the whole configuration.
  • -t/--template: Path to the template package to load. The templates are located in tergite_autocalibration.config.templates. If the autocompletion is installed for acli, then the templates should be shown as suggestions automatically.

Notes:

To run this command, please navigate to the root directory of the repository. The configuration package will be placed into the root directory, which is the default location for the application to detect the configuration package.

config save

Usage:

1
acli config save [OPTIONS]

Save the configuration.

Options:

  • -f/--filepath: Path to the configuration package to save. If the path name is ending with .zip, it will automatically create a zip file and treat it as if you are running with -z.
  • -z/--as-zip: Will make the configuration file be a zip archive.

bcc-export

Usage:

1
acli bcc-export [OPTIONS]

Create a calibration_seed.toml file that can be used from the backend to push calibration values to the database.

Options:

  • -q/--qubits: Qubit input e.g. "q00,q01,q02,q03,q04" or "q01-q05" or "q01-q06, q08". If the input is an integer e.g. 3, it will generate "q01,q02,q03".
  • -c/--couplers: Couplers to export e.g. "q00_q01" as comma-separated list
  • -o/--output-file: calibration_seed.toml to write.

config generate

Usage:

1
acli config generate [OPTIONS]

Save the configuration.

Options:

  • -h/--host: Host address where to run the interface of the generator. Default: 127.0.0.1
  • -p/--port: Port on which the application will serve the generator. Default: 8079

Dataset browser

browser

Starts the dataset browser.

Usage:

1
acli browser --datadir [OPTIONS]

Options:

  • --datadir PATH: Folder to take the plot data from
  • --liveplotting: Whether plots should be updated in real time (default: False)
  • --log-level INT: Log-level as in the Python logging package to be used in the logs (default: 30)

Redis handling

redis

Tools to work with the redis backend.

Usage:

1
acli redis save-file [FILENAME]

Store a backup of redis in .json format.

Usage:

1
acli redis load-file [FILENAME]

Load a backup of redis from a .json format. The file must follow the standard from the acli redis save-file function.

Joke Command

joke

Prints a random joke to lighten the mood.

Usage:

1
acli joke

This command fetches and displays a random joke, excluding potentially offensive content.

Notes

  • The CLI uses the Python library typer for command-line interface creation.
  • Some commands may require additional configuration or environment variables to be set.
  • When using the -r option for rerunning analysis, make sure to also specify the node name using -n.

For more detailed information about each command, use the --help option with any command or subcommand.