CLI Reference#

Adapt command structure: adapt <command> [options]

adapt run-nexrad   # run the processing pipeline
adapt config       # generate a config.yaml template
adapt dashboard    # open the GUI dashboard
adapt postprocess  # enrich an existing repository with extension tables

adapt --version prints the installed version and the path Adapt was installed to. adapt --help and adapt <command> --help print this same reference from the command line.

`adapt run-nexrad`#

Download and process NEXRAD Level-II data from AWS S3.

adapt run-nexrad [config.yaml] --radar KLOT --mode realtime
adapt run-nexrad [config.yaml] --radar KDIX --base-dir /data \
    --start-time 2025-03-05T15:00:00Z \
    --end-time   2025-03-05T18:00:00Z

Positional argument#

Argument	Description
`config`	Path to a config file (`.yaml` or `.py` with a `CONFIG` dict). Optional — expert defaults are used when omitted.

Options#

Flag	Default	Description
`--radar SITE`	—	4-letter NEXRAD site code, e.g. `KLOT`, `KDIX`, `KFTG`
`--mode`	`realtime`	`realtime` (continuous) or `historical` (fixed window)
`--start-time ISO`	—	Start of historical window, ISO 8601 (e.g. `2025-03-05T15:00:00Z`)
`--end-time ISO`	—	End of historical window, ISO 8601
`--base-dir PATH`	—	Root output directory for all artifacts
`--run-id ID`	—	Resume a previous run by ID (format: `YYYYMONDD-HHMM-RADAR`)
`--max-runtime MIN`	—	Stop after this many minutes (realtime mode only)
`--rerun`	off	Delete existing output for this radar before starting
`--no-plot`	off	Disable the plot consumer thread
`--plot-interval SEC`	`2.0`	How often the plot consumer checks for new data (seconds)
`--show-plots`	off	Open a live window showing plots as they are produced
`--only NODES`	—	Comma-separated pipeline node names to run; skip the rest. Mutually exclusive with `--not`
`--not NODES`	—	Comma-separated pipeline node names to skip; run the rest. Mutually exclusive with `--only`
`-v`, `--verbose`	off	Enable DEBUG-level logging

Mode selection logic#

If --start-time or --end-time is supplied, mode is automatically set to historical even if --mode is not given explicitly.

`adapt config`#

Write a commented YAML configuration template to disk.

adapt config                     # writes ./config.yaml
adapt config /path/to/my.yaml   # writes to a specific path

Argument	Description
`output`	Destination path. Defaults to `./config.yaml`. If a directory is given, `config.yaml` is written inside it.

Flag	Default	Description
`--pipeline NAME`	`nexrad`	Pipeline to generate a template for
`--extensions PATHS`	—	Comma-separated extension module import paths to include (e.g. `adapt.execution.nodes.cell_volume_stats`)

The generated file includes all tunable parameters with inline comments. Edit it, then pass it as the first argument to adapt run-nexrad.

`adapt postprocess`#

Enrich an existing repository with post-processing extension tables (e.g. associating lightning or other external observations with completed storm tracks).

adapt postprocess --module xlma_stat
adapt postprocess --repository /data/radar --module xlma_stat --input-dir /data/lma -v

Flag	Default	Description
`--repository PATH`	`.`	Repository to enrich
`--module MODULE [MODULE ...]`	—	Post-processing module(s) to run
`--input-dir PATH`	—	Directory containing the external dataset to associate
`--config PATH`	—	Config file with module-specific settings
`-v`, `--verbose`	off	Enable DEBUG-level logging

`adapt dashboard`#

Launch the read-only GUI dashboard.

adapt dashboard
adapt dashboard --repo ~/adapt_output

Flag	Description
`--repo PATH`	Pre-populate the repository path field in the GUI.

The dashboard reads from an existing output repository. It does not start or affect a running pipeline. Run it in a second terminal while the pipeline runs.

CLI Reference#

adapt run-nexrad#