enumeray model init
Generate a runnable local author starter with declarations, wrappers, tests and fixtures.
Use when: You need a concrete starting point for a new scientific model without an API or installed release.
Role and prerequisites: Model author. CLI installed; no credentials or API connection required. Both Python and executable starters need Python 3.9+ for the generated wrapper and local tests.
Example
Section titled “Example”enumeray model init scalar-double --directory ./scalar-author --kind python --io json --driver processcd scalar-authorpython3 -m unittest discover -s . -p 'test_starter.py'Use --kind executable for a wrapper around another language, --io files for a file-input/output starter, or --driver slurm for an installed batch-script starter. See wrappers and executable examples.
Inputs and options
Section titled “Inputs and options”| Input | Meaning and use |
|---|---|
<ID> |
Stable model ID embedded in the editable publication, such as scalar-double. |
--directory <DIRECTORY> |
New local author-project directory; any existing directory is refused. This creates starter files only and does not install or register a model. |
--kind <KIND> |
Starter wrapper language: python includes a Python runner, executable supplies a command-wrapper scaffold. Both need adaptation to your scientific code. |
--io <IO> |
Starter I/O contract: json for named scalar/array values, files for artifact input/output examples. It determines the generated declarations, fixtures and wrapper. |
--driver <DRIVER> |
Starter execution shape: process for direct process execution or slurm for an installed batch wrapper. Match it to the granted context. |
Every command also accepts the profile, API, JSON and timeout options.
Result and next step
Section titled “Result and next step”Creates the starter project and lists its files. Includes one directly editable draft publication.json. It does not create a catalogue model, install code or allocate a candidate ID.
Edit publication.json and the wrapper, test locally, then obtain a compute grant and install an immutable release.
Exact syntax, defaults and accepted values
Create a runnable author project locally, without credentials or installation
Usage: enumeray model init [OPTIONS] --directory <DIRECTORY> <ID>
Arguments: <ID> Stable model ID embedded in the generated publication source
Options: --api <API> User API origin override; credentials never cross origins. Host/runner commands use explicit configuration
--directory <DIRECTORY> New local starter project directory; any existing path is refused
--kind <KIND> Starter wrapper language; adapt the generated wrapper to your scientific code
Possible values: - python: Generate a Python model wrapper and local tests - executable: Generate an external executable wrapper and local tests
[default: python]
--profile <PROFILE> Named local user API/login profile; host/runner commands use explicit configuration
[default: default]
--config <CONFIG> Private user profiles file; defaults to $XDG_CONFIG_HOME/enumeray/profiles.json or ~/.config/enumeray/profiles.json
--io <IO> Starter input/output contract: named JSON values or declared file slots
Possible values: - json: Use named scalar and array JSON inputs and outputs - files: Include declared file inputs, outputs and local fixtures
[default: json]
--driver <DRIVER> Starter execution driver; Slurm requires site-specific batch/runtime preparation
Possible values: - process: Execute the installed wrapper as a direct process - slurm: Generate a batch wrapper for site-specific Slurm preparation
[default: process]
--json Emit user API results as JSON (NDJSON for following logs) and runtime errors on stderr; host/runner output follows its own protocol
--timeout <TIMEOUT> User API HTTP timeout in seconds; live logs use it for read inactivity. Does not limit execution time
[default: 60]
-h, --help Print help (see a summary with '-h')Documentation 205dcb3116ea · public sources 8c70356d340f · built .Download documentation identity.