eogrow.core.schemas
Module defining a base pipeline schema and custom fields
For each pipeline a separate schema has to be defined which inherits from PipelineSchema. Such schema should be placed as an internal class of the implemented pipeline class
- pydantic model eogrow.core.schemas.ManagerSchema[source]
Bases:
SchemaA basic schema for managers, to be used as a parent class for defining manager schemas
Create a new model by parsing and validating input data from keyword arguments.
Raises ValidationError if the input data cannot be parsed to form a valid model.
- Fields:
- field manager: str | None = None
An import path to this specific manager.
- pydantic model eogrow.core.schemas.PipelineSchema[source]
Bases:
SchemaBase schema of the Pipeline class.
Create a new model by parsing and validating input data from keyword arguments.
Raises ValidationError if the input data cannot be parsed to form a valid model.
- Fields:
- field area: ManagerSchema [Required]
A schema of an implementation of AreaManager class
- Validated by:
- field debug: bool = False
Run pipeline without the ray wrapper to enable debugging.
- field logging: ManagerSchema [Required]
A schema of an implementation of LoggingManager class
- Validated by:
- field pipeline: str | None = None
Import path to an implementation of Pipeline class.
- field pipeline_name: str | None = None
Custom pipeline name for easier identification in logs. By default the class name is used.
- field raise_on_failure: bool = True
Raise an exception if run_procedure returns some executions in failed.
- field raise_on_temporal_mismatch: bool = False
Treat TemporalDimensionWarning as an exception during EOExecution.
- field skip_existing: bool = False
Whether or not to skip already processed patches. In order to use this functionality a pipeline must implement the filter_patch_list method.
- field storage: ManagerSchema [Required]
A schema of an implementation of StorageManager class
- Validated by:
- field test_subset: List[int | str] | None = None
A list of EOPatch indices and/or names for which the pipeline is executed. Used for testing, can be set through CLI with the -t flag.
- field worker_resources: Dict[str, Any] [Optional]
Keyword arguments passed to ray tasks when executing via RayExecutor. The options are specified [here](https://docs.ray.io/en/latest/ray-core/api/doc/ray.remote_function.RemoteFunction.options.html).