fractal task
usage: fractal task [-h]
{list,make-core,make-not-core,collect,collect-custom,check-collection,new,edit}
...
Task commands.
options:
-h, --help show this help message and exit
Valid sub-commands:
{list,make-core,make-not-core,collect,collect-custom,check-collection,new,edit}
fractal task list
usage: fractal task list [-h]
List tasks.
options:
-h, --help show this help message and exit
fractal task make-core
usage: fractal task make-core [-h] task_ids [task_ids ...]
Mark tasks as core.
positional arguments:
task_ids List of the IDs of tasks that should be made core.
options:
-h, --help show this help message and exit
fractal task make-not-core
usage: fractal task make-not-core [-h] task_ids [task_ids ...]
Mark tasks as non-core.
positional arguments:
task_ids List of the IDs of tasks that should be made core.
options:
-h, --help show this help message and exit
fractal task collect
usage: fractal task collect [-h] [--python-version PYTHON_VERSION]
[--package-version PACKAGE_VERSION]
[--package-extras PACKAGE_EXTRAS]
[--pre-pinned-dependency PRE_PINNED_DEPENDENCY]
[--post-pinned-dependency POST_PINNED_DEPENDENCY]
[--private]
package
Install and collect all tasks from a pip-installable package.
positional arguments:
package Package name or absolute path to local package.
options:
-h, --help show this help message and exit
--python-version PYTHON_VERSION
Select the python version to use for this package.
--package-version PACKAGE_VERSION
Select the package version.
--package-extras PACKAGE_EXTRAS
Comma separated list of extra components for the
package to be installed, e.g., `collect fractal-tasks-
core --package-extras=torch,tensorflow` will trigger
the installation of `fractal-tasks-
core[torch,tensorflow]`.
--pre-pinned-dependency PRE_PINNED_DEPENDENCY
Package/version pair representing a pre-pinned-version
dependency, in the form `collect fractal-tasks-core
--pre-pinned-dependency pydantic=1.10.0`. Include
`--pre-pinned-dependency` multiple times to pin
several packages to specific versions.
--post-pinned-dependency POST_PINNED_DEPENDENCY
Package/version pair representing a post-pinned-
version dependency, in the form `collect fractal-
tasks-core --post-pinned-dependency pydantic=1.10.0`.
Include `--post-pinned-dependency` multiple times to
pin several packages to specific versions.
--private Make task group private.
fractal task collect-custom
usage: fractal task collect-custom [-h] --version VERSION
(--package-name PACKAGE_NAME | --package-root PACKAGE_ROOT)
[--private]
label python_interpreter manifest
Collect all tasks from a custom Python interpreter.
positional arguments:
label A common label identifying this package.
python_interpreter Absolute path to the Python interpreter to be used for
running tasks.
manifest Local path of the Manifest of the Fractal task
package.
options:
-h, --help show this help message and exit
--version VERSION Version of tasks to be collected.
--package-name PACKAGE_NAME
Name of the package, as used in 'import
<package_name>'; this is then used to extract the
package directory (package_root) via
'importlib.util.find_spec <package_name>'.
--package-root PACKAGE_ROOT
The folder where the package is installed. If not
provided, it will be automatically inferred based on
'package_name'.
--private Make task group private.
fractal task check-collection
usage: fractal task check-collection [-h] [--include-logs]
task_group_activity_id
Check status of background task collection processes.
positional arguments:
task_group_activity_id
Activity ID of the collection (see output of `task
collect`).
options:
-h, --help show this help message and exit
--include-logs Also include task-collection logs.
fractal task new
usage: fractal task new [-h] [--task-type TASK_TYPE]
[--command-non-parallel COMMAND_NON_PARALLEL]
[--command-parallel COMMAND_PARALLEL] --version
VERSION [--meta-non-parallel META_NON_PARALLEL]
[--meta-parallel META_PARALLEL]
[--args-schema-non-parallel ARGS_SCHEMA_NON_PARALLEL]
[--args-schema-parallel ARGS_SCHEMA_PARALLEL]
[--args-schema-version ARGS_SCHEMA_VERSION]
[--private]
name
Create new task.
positional arguments:
name A human readable name for the task.
options:
-h, --help show this help message and exit
--task-type TASK_TYPE
The task type (e.g. 'parallel', 'non_parallel',
'compound').
--command-non-parallel COMMAND_NON_PARALLEL
The non parallel command that executes the task.
--command-parallel COMMAND_PARALLEL
The parallel command that executes the task.
--version VERSION Task version.
--meta-non-parallel META_NON_PARALLEL
Path to JSON file with meta non parallel arguments.
--meta-parallel META_PARALLEL
Path to JSON file with meta parallel arguments.
--args-schema-non-parallel ARGS_SCHEMA_NON_PARALLEL
Path to JSON file with args non parallel arguments.
--args-schema-parallel ARGS_SCHEMA_PARALLEL
Path to JSON file with arg parallel arguments.
--args-schema-version ARGS_SCHEMA_VERSION
Label encoding how the task-arguments JSON Schema was
generated (e.g. `pydantic_v1`).
--private Make task group private.
fractal task edit
usage: fractal task edit [-h] (--id ID | --name NAME) [--version VERSION]
[--command-non-parallel COMMAND_NON_PARALLEL]
[--command-parallel COMMAND_PARALLEL]
[--input-types INPUT_TYPES]
[--output-types OUTPUT_TYPES]
Edit task.
options:
-h, --help show this help message and exit
--id ID ID of the task to edit.
--name NAME Name of the task to edit.
--version VERSION Version of the task to edit (only accepted in
combination with `--name`).
--command-non-parallel COMMAND_NON_PARALLEL
New task non parallel command.
--command-parallel COMMAND_PARALLEL
New task parallel command.
--input-types INPUT_TYPES
Path to JSON file with new input types.
--output-types OUTPUT_TYPES
Path to JSON file with new output types.