Databricks

zenml.integrations.databricks

Initialization of the Databricks integration for ZenML.

Attributes

DATABRICKS = 'databricks' module-attribute

DATABRICKS_MODEL_DEPLOYER_FLAVOR = 'databricks' module-attribute

DATABRICKS_ORCHESTRATOR_FLAVOR = 'databricks' module-attribute

DATABRICKS_SERVICE_ARTIFACT = 'databricks_deployment_service' module-attribute

DATABRICKS_STEP_OPERATOR_FLAVOR = 'databricks' module-attribute

Classes

DatabricksIntegration

Bases: Integration

Definition of Databricks Integration for ZenML.

Functions

flavors() -> List[Type[Flavor]] classmethod

Declare the stack component flavors for the Databricks integration.

Returns:

Type	Description
List[Type[Flavor]]	List of stack component flavors for this integration.

Source code in src/zenml/integrations/databricks/__init__.py

@classmethod
def flavors(cls) -> List[Type[Flavor]]:
    """Declare the stack component flavors for the Databricks integration.

    Returns:
        List of stack component flavors for this integration.
    """
    from zenml.integrations.databricks.flavors import (
        DatabricksModelDeployerFlavor,
        DatabricksOrchestratorFlavor,
        DatabricksStepOperatorFlavor,
    )

    return [
        DatabricksOrchestratorFlavor,
        DatabricksStepOperatorFlavor,
        DatabricksModelDeployerFlavor,
    ]

get_requirements(target_os: Optional[str] = None, python_version: Optional[str] = None) -> List[str] classmethod

Method to get the requirements for the integration.

Parameters:

Name	Type	Description	Default
target_os	Optional[str]	The target operating system to get the requirements for.	None
python_version	Optional[str]	The Python version to use for the requirements.	None

Returns:

Type	Description
List[str]	A list of requirements.

Source code in src/zenml/integrations/databricks/__init__.py

@classmethod
def get_requirements(
    cls,
    target_os: Optional[str] = None,
    python_version: Optional[str] = None,
) -> List[str]:
    """Method to get the requirements for the integration.

    Args:
        target_os: The target operating system to get the requirements for.
        python_version: The Python version to use for the requirements.

    Returns:
        A list of requirements.
    """
    from zenml.integrations.numpy import NumpyIntegration
    from zenml.integrations.pandas import PandasIntegration

    return (
        cls.REQUIREMENTS
        + NumpyIntegration.get_requirements(
            target_os=target_os, python_version=python_version
        )
        + PandasIntegration.get_requirements(
            target_os=target_os, python_version=python_version
        )
    )

Flavor

Class for ZenML Flavors.

Attributes

config_class: Type[StackComponentConfig] abstractmethod property

Returns StackComponentConfig config class.

Returns:

Type	Description
Type[StackComponentConfig]	The config class.

config_schema: Dict[str, Any] property

The config schema for a flavor.

Returns:

Type	Description
Dict[str, Any]	The config schema.

display_name: Optional[str] property

The display name of the flavor.

By default, converts the technical name to a human-readable format. For example, "vm_kubernetes" becomes "VM Kubernetes". Flavors can override this to provide custom display names.

Returns:

Type	Description
Optional[str]	The display name of the flavor.

docs_url: Optional[str] property

A url to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs url.

implementation_class: Type[StackComponent] abstractmethod property

Implementation class for this flavor.

Returns:

Type	Description
Type[StackComponent]	The implementation class for this flavor.

logo_url: Optional[str] property

A url to represent the flavor in the dashboard.

Returns:

Type	Description
Optional[str]	The flavor logo.

name: str abstractmethod property

The flavor name.

Returns:

Type	Description
str	The flavor name.

sdk_docs_url: Optional[str] property

A url to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs url.

service_connector_requirements: Optional[ServiceConnectorRequirements] property

Service connector resource requirements for service connectors.

Specifies resource requirements that are used to filter the available service connector types that are compatible with this flavor.

Returns:

Type	Description
Optional[ServiceConnectorRequirements]	Requirements for compatible service connectors, if a service
Optional[ServiceConnectorRequirements]	connector is required for this flavor.

type: StackComponentType abstractmethod property

The stack component type.

Returns:

Type	Description
StackComponentType	The stack component type.

Functions

from_model(flavor_model: FlavorResponse) -> Flavor classmethod

Loads a flavor from a model.

Parameters:

Name	Type	Description	Default
flavor_model	FlavorResponse	The model to load from.	required

Raises:

Type	Description
CustomFlavorImportError	If the custom flavor can't be imported.
ImportError	If the flavor can't be imported.

Returns:

Type	Description
Flavor	The loaded flavor.

Source code in src/zenml/stack/flavor.py

@classmethod
def from_model(cls, flavor_model: FlavorResponse) -> "Flavor":
    """Loads a flavor from a model.

    Args:
        flavor_model: The model to load from.

    Raises:
        CustomFlavorImportError: If the custom flavor can't be imported.
        ImportError: If the flavor can't be imported.

    Returns:
        The loaded flavor.
    """
    try:
        flavor = source_utils.load(flavor_model.source)()
    except (ModuleNotFoundError, ImportError, NotImplementedError) as err:
        if flavor_model.is_custom:
            flavor_module, _ = flavor_model.source.rsplit(".", maxsplit=1)
            expected_file_path = os.path.join(
                source_utils.get_source_root(),
                flavor_module.replace(".", os.path.sep),
            )
            raise CustomFlavorImportError(
                f"Couldn't import custom flavor {flavor_model.name}: "
                f"{err}. Make sure the custom flavor class "
                f"`{flavor_model.source}` is importable. If it is part of "
                "a library, make sure it is installed. If "
                "it is a local code file, make sure it exists at "
                f"`{expected_file_path}.py`."
            )
        else:
            raise ImportError(
                f"Couldn't import flavor {flavor_model.name}: {err}"
            )
    return cast(Flavor, flavor)

generate_default_docs_url() -> str

Generate the doc urls for all inbuilt and integration flavors.

Note that this method is not going to be useful for custom flavors, which do not have any docs in the main zenml docs.

Returns:

Type	Description
str	The complete url to the zenml documentation

Source code in src/zenml/stack/flavor.py

def generate_default_docs_url(self) -> str:
    """Generate the doc urls for all inbuilt and integration flavors.

    Note that this method is not going to be useful for custom flavors,
    which do not have any docs in the main zenml docs.

    Returns:
        The complete url to the zenml documentation
    """
    component_type = self.type.plural.replace("_", "-")
    name = self.name.replace("_", "-")

    base = "https://docs.zenml.io"
    return f"{base}/stack-components/{component_type}/{name}"

generate_default_sdk_docs_url() -> str

Generate SDK docs url for a flavor.

Returns:

Type	Description
str	The complete url to the zenml SDK docs

Source code in src/zenml/stack/flavor.py

def generate_default_sdk_docs_url(self) -> str:
    """Generate SDK docs url for a flavor.

    Returns:
        The complete url to the zenml SDK docs
    """
    from zenml import __version__

    base = f"https://sdkdocs.zenml.io/{__version__}"

    component_type = self.type.plural

    if "zenml.integrations" in self.__module__:
        # Get integration name out of module path which will look something
        #  like this "zenml.integrations.<integration>....
        integration = self.__module__.split(
            "zenml.integrations.", maxsplit=1
        )[1].split(".")[0]

        # Get the config class name to point to the specific class
        config_class_name = self.config_class.__name__

        return (
            f"{base}/integration_code_docs"
            f"/integrations-{integration}"
            f"#zenml.integrations.{integration}.flavors.{config_class_name}"
        )

    else:
        return (
            f"{base}/core_code_docs/core-{component_type}/"
            f"#{self.__module__}"
        )

to_model(integration: Optional[str] = None, is_custom: bool = True) -> FlavorRequest

Converts a flavor to a model.

Parameters:

Name	Type	Description	Default
integration	Optional[str]	The integration to use for the model.	None
is_custom	bool	Whether the flavor is a custom flavor.	True

Returns:

Type	Description
FlavorRequest	The model.

Source code in src/zenml/stack/flavor.py

def to_model(
    self,
    integration: Optional[str] = None,
    is_custom: bool = True,
) -> FlavorRequest:
    """Converts a flavor to a model.

    Args:
        integration: The integration to use for the model.
        is_custom: Whether the flavor is a custom flavor.

    Returns:
        The model.
    """
    connector_requirements = self.service_connector_requirements
    connector_type = (
        connector_requirements.connector_type
        if connector_requirements
        else None
    )
    resource_type = (
        connector_requirements.resource_type
        if connector_requirements
        else None
    )
    resource_id_attr = (
        connector_requirements.resource_id_attr
        if connector_requirements
        else None
    )

    model = FlavorRequest(
        name=self.name,
        display_name=self.display_name,
        type=self.type,
        source=source_utils.resolve(self.__class__).import_path,
        config_schema=self.config_schema,
        connector_type=connector_type,
        connector_resource_type=resource_type,
        connector_resource_id_attr=resource_id_attr,
        integration=integration,
        logo_url=self.logo_url,
        docs_url=self.docs_url,
        sdk_docs_url=self.sdk_docs_url,
        is_custom=is_custom,
    )
    return model

Integration

Base class for integration in ZenML.

Functions

activate() -> None classmethod

Abstract method to activate the integration.

Source code in src/zenml/integrations/integration.py

@classmethod
def activate(cls) -> None:
    """Abstract method to activate the integration."""

check_installation() -> bool classmethod

Method to check whether the required packages are installed.

Returns:

Type	Description
bool	True if all required packages are installed, False otherwise.

Source code in src/zenml/integrations/integration.py

@classmethod
def check_installation(cls) -> bool:
    """Method to check whether the required packages are installed.

    Returns:
        True if all required packages are installed, False otherwise.
    """
    for requirement in cls.get_requirements():
        parsed_requirement = Requirement(requirement)

        if not requirement_installed(parsed_requirement):
            logger.debug(
                "Requirement '%s' for integration '%s' is not installed "
                "or installed with the wrong version.",
                requirement,
                cls.NAME,
            )
            return False

        dependencies = get_dependencies(parsed_requirement)

        for dependency in dependencies:
            if not requirement_installed(dependency):
                logger.debug(
                    "Requirement '%s' for integration '%s' is not "
                    "installed or installed with the wrong version.",
                    dependency,
                    cls.NAME,
                )
                return False

    logger.debug(
        f"Integration '{cls.NAME}' is installed correctly with "
        f"requirements {cls.get_requirements()}."
    )
    return True

flavors() -> List[Type[Flavor]] classmethod

Abstract method to declare new stack component flavors.

Returns:

Type	Description
List[Type[Flavor]]	A list of new stack component flavors.

Source code in src/zenml/integrations/integration.py

@classmethod
def flavors(cls) -> List[Type[Flavor]]:
    """Abstract method to declare new stack component flavors.

    Returns:
        A list of new stack component flavors.
    """
    return []

get_requirements(target_os: Optional[str] = None, python_version: Optional[str] = None) -> List[str] classmethod

Method to get the requirements for the integration.

Parameters:

Name	Type	Description	Default
target_os	Optional[str]	The target operating system to get the requirements for.	None
python_version	Optional[str]	The Python version to use for the requirements.	None

Returns:

Type	Description
List[str]	A list of requirements.

Source code in src/zenml/integrations/integration.py

@classmethod
def get_requirements(
    cls,
    target_os: Optional[str] = None,
    python_version: Optional[str] = None,
) -> List[str]:
    """Method to get the requirements for the integration.

    Args:
        target_os: The target operating system to get the requirements for.
        python_version: The Python version to use for the requirements.

    Returns:
        A list of requirements.
    """
    return cls.REQUIREMENTS

get_uninstall_requirements(target_os: Optional[str] = None) -> List[str] classmethod

Method to get the uninstall requirements for the integration.

Parameters:

Name	Type	Description	Default
target_os	Optional[str]	The target operating system to get the requirements for.	None

Returns:

Type	Description
List[str]	A list of requirements.

Source code in src/zenml/integrations/integration.py

@classmethod
def get_uninstall_requirements(
    cls, target_os: Optional[str] = None
) -> List[str]:
    """Method to get the uninstall requirements for the integration.

    Args:
        target_os: The target operating system to get the requirements for.

    Returns:
        A list of requirements.
    """
    ret = []
    for each in cls.get_requirements(target_os=target_os):
        is_ignored = False
        for ignored in cls.REQUIREMENTS_IGNORED_ON_UNINSTALL:
            if each.startswith(ignored):
                is_ignored = True
                break
        if not is_ignored:
            ret.append(each)
    return ret

Modules

flavors

Databricks integration flavors.

Classes

DatabricksAccessControlRequest

Bases: BaseModel

Databricks job access control entry.

DatabricksAvailabilityType

Bases: StrEnum

Databricks availability type.

DatabricksModelDeployerConfig(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseModelDeployerConfig

Configuration for the Databricks model deployer.

Attributes:

Name	Type	Description
host	str	Databricks host.
secret_name	Optional[str]	Secret name to use for authentication.
client_id	Optional[str]	Databricks client id.
client_secret	Optional[str]	Databricks client secret.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references don't clash with pydantic validation.

    StackComponents allow the specification of all their string attributes
    using secret references of the form `{{secret_name.key}}`. This however
    is only possible when the stack component does not perform any explicit
    validation of this attribute using pydantic validators. If this were
    the case, the validation would run on the secret reference and would
    fail or in the worst case, modify the secret reference and lead to
    unexpected behavior. This method ensures that no attributes that require
    custom pydantic validation are set as secret references.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using
            plain-text secrets.
        **kwargs: Arguments to initialize this stack component.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            is passed as a secret reference, or if the `name` attribute
            was passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}` for a `{self.__class__.__name__}` "
                    "stack component. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure your stack "
                    "components with secrets here: "
                    "https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if pydantic_utils.has_validators(
            pydantic_class=self.__class__, field_name=key
        ):
            raise ValueError(
                f"Passing the stack component attribute `{key}` as a "
                "secret reference is not allowed as additional validation "
                "is required for this attribute."
            )

    super().__init__(**kwargs)

DatabricksModelDeployerFlavor

Bases: BaseModelDeployerFlavor

Databricks Endpoint model deployer flavor.

Attributes

config_class: Type[DatabricksModelDeployerConfig] property

Returns DatabricksModelDeployerConfig config class.

Returns:

Type	Description
Type[DatabricksModelDeployerConfig]	The config class.

docs_url: Optional[str] property

A url to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs url.

implementation_class: Type[DatabricksModelDeployer] property

Implementation class for this flavor.

Returns:

Type	Description
Type[DatabricksModelDeployer]	The implementation class.

logo_url: str property

A url to represent the flavor in the dashboard.

Returns:

Type	Description
str	The flavor logo.

name: str property

Name of the flavor.

Returns:

Type	Description
str	The name of the flavor.

sdk_docs_url: Optional[str] property

A url to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs url.

DatabricksOrchestratorConfig(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseOrchestratorConfig, DatabricksOrchestratorSettings

Databricks orchestrator base config.

Attributes:

Name	Type	Description
host	str	Databricks host.
client_id	Optional[str]	Databricks client id.
client_secret	Optional[str]	Databricks client secret.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references don't clash with pydantic validation.

    StackComponents allow the specification of all their string attributes
    using secret references of the form `{{secret_name.key}}`. This however
    is only possible when the stack component does not perform any explicit
    validation of this attribute using pydantic validators. If this were
    the case, the validation would run on the secret reference and would
    fail or in the worst case, modify the secret reference and lead to
    unexpected behavior. This method ensures that no attributes that require
    custom pydantic validation are set as secret references.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using
            plain-text secrets.
        **kwargs: Arguments to initialize this stack component.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            is passed as a secret reference, or if the `name` attribute
            was passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}` for a `{self.__class__.__name__}` "
                    "stack component. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure your stack "
                    "components with secrets here: "
                    "https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if pydantic_utils.has_validators(
            pydantic_class=self.__class__, field_name=key
        ):
            raise ValueError(
                f"Passing the stack component attribute `{key}` as a "
                "secret reference is not allowed as additional validation "
                "is required for this attribute."
            )

    super().__init__(**kwargs)

Attributes

is_local: bool property

Checks if this stack component is running locally.

Returns:

Type	Description
bool	True if this config is for a local component, False otherwise.

is_remote: bool property

Checks if this stack component is running remotely.

Returns:

Type	Description
bool	True if this config is for a remote component, False otherwise.

is_schedulable: bool property

Whether the orchestrator is schedulable or not.

Returns:

Type	Description
bool	Whether the orchestrator is schedulable or not.

DatabricksOrchestratorFlavor

Bases: BaseOrchestratorFlavor

Databricks orchestrator flavor.

Attributes

config_class: Type[DatabricksOrchestratorConfig] property

Returns DatabricksOrchestratorConfig config class.

Returns:

Type	Description
Type[DatabricksOrchestratorConfig]	The config class.

docs_url: Optional[str] property

A url to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs url.

implementation_class: Type[DatabricksOrchestrator] property

Implementation class for this flavor.

Returns:

Type	Description
Type[DatabricksOrchestrator]	The implementation class.

logo_url: str property

A url to represent the flavor in the dashboard.

Returns:

Type	Description
str	The flavor logo.

name: str property

Name of the flavor.

Returns:

Type	Description
str	The name of the flavor.

sdk_docs_url: Optional[str] property

A url to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs url.

DatabricksOrchestratorSettings(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: DatabricksBaseSettings

Databricks orchestrator settings.

Source code in src/zenml/config/secret_reference_mixin.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references are only passed for valid fields.

    This method ensures that secret references are not passed for fields
    that explicitly prevent them or require pydantic validation.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using plain-text secrets.
        **kwargs: Arguments to initialize this object.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            or an attribute which explicitly disallows secret references
            is passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}`. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure values with secrets "
                    "here: https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if secret_utils.is_clear_text_field(field):
            raise ValueError(
                f"Passing the `{key}` attribute as a secret reference is "
                "not allowed."
            )

        requires_validation = has_validators(
            pydantic_class=self.__class__, field_name=key
        )
        if requires_validation:
            raise ValueError(
                f"Passing the attribute `{key}` as a secret reference is "
                "not allowed as additional validation is required for "
                "this attribute."
            )

    super().__init__(**kwargs)

DatabricksPermissionLevel

Bases: StrEnum

Databricks job permission levels.

DatabricksStepOperatorConfig(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseStepOperatorConfig, DatabricksStepOperatorSettings

Databricks step operator config.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references don't clash with pydantic validation.

    StackComponents allow the specification of all their string attributes
    using secret references of the form `{{secret_name.key}}`. This however
    is only possible when the stack component does not perform any explicit
    validation of this attribute using pydantic validators. If this were
    the case, the validation would run on the secret reference and would
    fail or in the worst case, modify the secret reference and lead to
    unexpected behavior. This method ensures that no attributes that require
    custom pydantic validation are set as secret references.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using
            plain-text secrets.
        **kwargs: Arguments to initialize this stack component.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            is passed as a secret reference, or if the `name` attribute
            was passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}` for a `{self.__class__.__name__}` "
                    "stack component. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure your stack "
                    "components with secrets here: "
                    "https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if pydantic_utils.has_validators(
            pydantic_class=self.__class__, field_name=key
        ):
            raise ValueError(
                f"Passing the stack component attribute `{key}` as a "
                "secret reference is not allowed as additional validation "
                "is required for this attribute."
            )

    super().__init__(**kwargs)

Attributes

is_local: bool property

Checks if this stack component is running locally.

Returns:

Type	Description
bool	False, because the Databricks step operator always runs remotely.

is_remote: bool property

Checks if this stack component is running remotely.

Returns:

Type	Description
bool	True, because the Databricks step operator always runs remotely.

DatabricksStepOperatorFlavor

Bases: BaseStepOperatorFlavor

Databricks step operator flavor.

Attributes

config_class: Type[DatabricksStepOperatorConfig] property

Returns DatabricksStepOperatorConfig config class.

Returns:

Type	Description
Type[DatabricksStepOperatorConfig]	The config class.

docs_url: Optional[str] property

A URL to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs URL.

implementation_class: Type[DatabricksStepOperator] property

Implementation class for this flavor.

Returns:

Type	Description
Type[DatabricksStepOperator]	The implementation class.

logo_url: str property

A URL to represent the flavor in the dashboard.

Returns:

Type	Description
str	The flavor logo.

name: str property

Name of the flavor.

Returns:

Type	Description
str	The name of the flavor.

sdk_docs_url: Optional[str] property

A URL to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs URL.

DatabricksStepOperatorSettings(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: DatabricksBaseSettings

Databricks step operator settings.

Source code in src/zenml/config/secret_reference_mixin.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references are only passed for valid fields.

    This method ensures that secret references are not passed for fields
    that explicitly prevent them or require pydantic validation.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using plain-text secrets.
        **kwargs: Arguments to initialize this object.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            or an attribute which explicitly disallows secret references
            is passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}`. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure values with secrets "
                    "here: https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if secret_utils.is_clear_text_field(field):
            raise ValueError(
                f"Passing the `{key}` attribute as a secret reference is "
                "not allowed."
            )

        requires_validation = has_validators(
            pydantic_class=self.__class__, field_name=key
        )
        if requires_validation:
            raise ValueError(
                f"Passing the attribute `{key}` as a secret reference is "
                "not allowed as additional validation is required for "
                "this attribute."
            )

    super().__init__(**kwargs)

Modules

databricks_model_deployer_flavor

Databricks model deployer flavor.

Classes

DatabricksBaseConfig

Bases: BaseModel

Databricks Inference Endpoint configuration.

DatabricksModelDeployerConfig(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseModelDeployerConfig

Configuration for the Databricks model deployer.

Attributes:

Name	Type	Description
host	str	Databricks host.
secret_name	Optional[str]	Secret name to use for authentication.
client_id	Optional[str]	Databricks client id.
client_secret	Optional[str]	Databricks client secret.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references don't clash with pydantic validation.

    StackComponents allow the specification of all their string attributes
    using secret references of the form `{{secret_name.key}}`. This however
    is only possible when the stack component does not perform any explicit
    validation of this attribute using pydantic validators. If this were
    the case, the validation would run on the secret reference and would
    fail or in the worst case, modify the secret reference and lead to
    unexpected behavior. This method ensures that no attributes that require
    custom pydantic validation are set as secret references.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using
            plain-text secrets.
        **kwargs: Arguments to initialize this stack component.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            is passed as a secret reference, or if the `name` attribute
            was passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}` for a `{self.__class__.__name__}` "
                    "stack component. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure your stack "
                    "components with secrets here: "
                    "https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if pydantic_utils.has_validators(
            pydantic_class=self.__class__, field_name=key
        ):
            raise ValueError(
                f"Passing the stack component attribute `{key}` as a "
                "secret reference is not allowed as additional validation "
                "is required for this attribute."
            )

    super().__init__(**kwargs)

DatabricksModelDeployerFlavor

Bases: BaseModelDeployerFlavor

Databricks Endpoint model deployer flavor.

Attributes

config_class: Type[DatabricksModelDeployerConfig] property

Returns DatabricksModelDeployerConfig config class.

Returns:

Type	Description
Type[DatabricksModelDeployerConfig]	The config class.

docs_url: Optional[str] property

A url to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs url.

implementation_class: Type[DatabricksModelDeployer] property

Implementation class for this flavor.

Returns:

Type	Description
Type[DatabricksModelDeployer]	The implementation class.

logo_url: str property

A url to represent the flavor in the dashboard.

Returns:

Type	Description
str	The flavor logo.

name: str property

Name of the flavor.

Returns:

Type	Description
str	The name of the flavor.

sdk_docs_url: Optional[str] property

A url to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs url.

Functions

databricks_orchestrator_flavor

Databricks orchestrator base config and settings.

Classes

DatabricksOrchestratorConfig(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseOrchestratorConfig, DatabricksOrchestratorSettings

Databricks orchestrator base config.

Attributes:

Name	Type	Description
host	str	Databricks host.
client_id	Optional[str]	Databricks client id.
client_secret	Optional[str]	Databricks client secret.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references don't clash with pydantic validation.

    StackComponents allow the specification of all their string attributes
    using secret references of the form `{{secret_name.key}}`. This however
    is only possible when the stack component does not perform any explicit
    validation of this attribute using pydantic validators. If this were
    the case, the validation would run on the secret reference and would
    fail or in the worst case, modify the secret reference and lead to
    unexpected behavior. This method ensures that no attributes that require
    custom pydantic validation are set as secret references.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using
            plain-text secrets.
        **kwargs: Arguments to initialize this stack component.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            is passed as a secret reference, or if the `name` attribute
            was passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}` for a `{self.__class__.__name__}` "
                    "stack component. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure your stack "
                    "components with secrets here: "
                    "https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if pydantic_utils.has_validators(
            pydantic_class=self.__class__, field_name=key
        ):
            raise ValueError(
                f"Passing the stack component attribute `{key}` as a "
                "secret reference is not allowed as additional validation "
                "is required for this attribute."
            )

    super().__init__(**kwargs)

Attributes

is_local: bool property

Checks if this stack component is running locally.

Returns:

Type	Description
bool	True if this config is for a local component, False otherwise.

is_remote: bool property

Checks if this stack component is running remotely.

Returns:

Type	Description
bool	True if this config is for a remote component, False otherwise.

is_schedulable: bool property

Whether the orchestrator is schedulable or not.

Returns:

Type	Description
bool	Whether the orchestrator is schedulable or not.

DatabricksOrchestratorFlavor

Bases: BaseOrchestratorFlavor

Databricks orchestrator flavor.

Attributes

config_class: Type[DatabricksOrchestratorConfig] property

Returns DatabricksOrchestratorConfig config class.

Returns:

Type	Description
Type[DatabricksOrchestratorConfig]	The config class.

docs_url: Optional[str] property

A url to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs url.

implementation_class: Type[DatabricksOrchestrator] property

Implementation class for this flavor.

Returns:

Type	Description
Type[DatabricksOrchestrator]	The implementation class.

logo_url: str property

A url to represent the flavor in the dashboard.

Returns:

Type	Description
str	The flavor logo.

name: str property

Name of the flavor.

Returns:

Type	Description
str	The name of the flavor.

sdk_docs_url: Optional[str] property

A url to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs url.

DatabricksOrchestratorSettings(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: DatabricksBaseSettings

Databricks orchestrator settings.

Source code in src/zenml/config/secret_reference_mixin.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references are only passed for valid fields.

    This method ensures that secret references are not passed for fields
    that explicitly prevent them or require pydantic validation.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using plain-text secrets.
        **kwargs: Arguments to initialize this object.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            or an attribute which explicitly disallows secret references
            is passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}`. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure values with secrets "
                    "here: https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if secret_utils.is_clear_text_field(field):
            raise ValueError(
                f"Passing the `{key}` attribute as a secret reference is "
                "not allowed."
            )

        requires_validation = has_validators(
            pydantic_class=self.__class__, field_name=key
        )
        if requires_validation:
            raise ValueError(
                f"Passing the attribute `{key}` as a secret reference is "
                "not allowed as additional validation is required for "
                "this attribute."
            )

    super().__init__(**kwargs)

Functions

databricks_shared_settings

Shared Databricks settings models.

Classes

DatabricksAccessControlRequest

Bases: BaseModel

Databricks job access control entry.

DatabricksAvailabilityType

Bases: StrEnum

Databricks availability type.

DatabricksBaseSettings(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseSettings

Databricks execution settings shared by orchestrator and step operator.

Source code in src/zenml/config/secret_reference_mixin.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references are only passed for valid fields.

    This method ensures that secret references are not passed for fields
    that explicitly prevent them or require pydantic validation.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using plain-text secrets.
        **kwargs: Arguments to initialize this object.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            or an attribute which explicitly disallows secret references
            is passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}`. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure values with secrets "
                    "here: https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if secret_utils.is_clear_text_field(field):
            raise ValueError(
                f"Passing the `{key}` attribute as a secret reference is "
                "not allowed."
            )

        requires_validation = has_validators(
            pydantic_class=self.__class__, field_name=key
        )
        if requires_validation:
            raise ValueError(
                f"Passing the attribute `{key}` as a secret reference is "
                "not allowed as additional validation is required for "
                "this attribute."
            )

    super().__init__(**kwargs)

DatabricksPermissionLevel

Bases: StrEnum

Databricks job permission levels.

Functions

databricks_step_operator_flavor

Databricks step operator flavor.

Classes

DatabricksStepOperatorConfig(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: BaseStepOperatorConfig, DatabricksStepOperatorSettings

Databricks step operator config.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references don't clash with pydantic validation.

    StackComponents allow the specification of all their string attributes
    using secret references of the form `{{secret_name.key}}`. This however
    is only possible when the stack component does not perform any explicit
    validation of this attribute using pydantic validators. If this were
    the case, the validation would run on the secret reference and would
    fail or in the worst case, modify the secret reference and lead to
    unexpected behavior. This method ensures that no attributes that require
    custom pydantic validation are set as secret references.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using
            plain-text secrets.
        **kwargs: Arguments to initialize this stack component.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            is passed as a secret reference, or if the `name` attribute
            was passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}` for a `{self.__class__.__name__}` "
                    "stack component. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure your stack "
                    "components with secrets here: "
                    "https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if pydantic_utils.has_validators(
            pydantic_class=self.__class__, field_name=key
        ):
            raise ValueError(
                f"Passing the stack component attribute `{key}` as a "
                "secret reference is not allowed as additional validation "
                "is required for this attribute."
            )

    super().__init__(**kwargs)

Attributes

is_local: bool property

Checks if this stack component is running locally.

Returns:

Type	Description
bool	False, because the Databricks step operator always runs remotely.

is_remote: bool property

Checks if this stack component is running remotely.

Returns:

Type	Description
bool	True, because the Databricks step operator always runs remotely.

DatabricksStepOperatorFlavor

Bases: BaseStepOperatorFlavor

Databricks step operator flavor.

Attributes

config_class: Type[DatabricksStepOperatorConfig] property

Returns DatabricksStepOperatorConfig config class.

Returns:

Type	Description
Type[DatabricksStepOperatorConfig]	The config class.

docs_url: Optional[str] property

A URL to point at docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor docs URL.

implementation_class: Type[DatabricksStepOperator] property

Implementation class for this flavor.

Returns:

Type	Description
Type[DatabricksStepOperator]	The implementation class.

logo_url: str property

A URL to represent the flavor in the dashboard.

Returns:

Type	Description
str	The flavor logo.

name: str property

Name of the flavor.

Returns:

Type	Description
str	The name of the flavor.

sdk_docs_url: Optional[str] property

A URL to point at SDK docs explaining this flavor.

Returns:

Type	Description
Optional[str]	A flavor SDK docs URL.

DatabricksStepOperatorSettings(warn_about_plain_text_secrets: bool = False, **kwargs: Any)

Bases: DatabricksBaseSettings

Databricks step operator settings.

Source code in src/zenml/config/secret_reference_mixin.py

def __init__(
    self, warn_about_plain_text_secrets: bool = False, **kwargs: Any
) -> None:
    """Ensures that secret references are only passed for valid fields.

    This method ensures that secret references are not passed for fields
    that explicitly prevent them or require pydantic validation.

    Args:
        warn_about_plain_text_secrets: If true, then warns about using plain-text secrets.
        **kwargs: Arguments to initialize this object.

    Raises:
        ValueError: If an attribute that requires custom pydantic validation
            or an attribute which explicitly disallows secret references
            is passed as a secret reference.
    """
    for key, value in kwargs.items():
        try:
            field = self.__class__.model_fields[key]
        except KeyError:
            # Value for a private attribute or non-existing field, this
            # will fail during the upcoming pydantic validation
            continue

        if value is None:
            continue

        if not secret_utils.is_secret_reference(value):
            if (
                secret_utils.is_secret_field(field)
                and warn_about_plain_text_secrets
            ):
                logger.warning(
                    "You specified a plain-text value for the sensitive "
                    f"attribute `{key}`. This is currently only a warning, "
                    "but future versions of ZenML will require you to pass "
                    "in sensitive information as secrets. Check out the "
                    "documentation on how to configure values with secrets "
                    "here: https://docs.zenml.io/deploying-zenml/deploying-zenml/secret-management"
                )
            continue

        if secret_utils.is_clear_text_field(field):
            raise ValueError(
                f"Passing the `{key}` attribute as a secret reference is "
                "not allowed."
            )

        requires_validation = has_validators(
            pydantic_class=self.__class__, field_name=key
        )
        if requires_validation:
            raise ValueError(
                f"Passing the attribute `{key}` as a secret reference is "
                "not allowed as additional validation is required for "
                "this attribute."
            )

    super().__init__(**kwargs)

Functions

model_deployers

Initialization of the Databricks model deployers.

Classes

DatabricksModelDeployer(name: str, id: UUID, config: StackComponentConfig, flavor: str, type: StackComponentType, user: Optional[UUID], created: datetime, updated: datetime, environment: Optional[Dict[str, str]] = None, secrets: Optional[List[UUID]] = None, labels: Optional[Dict[str, Any]] = None, connector_requirements: Optional[ServiceConnectorRequirements] = None, connector: Optional[UUID] = None, connector_resource_id: Optional[str] = None, *args: Any, **kwargs: Any)

Bases: BaseModelDeployer

Databricks endpoint model deployer.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self,
    name: str,
    id: UUID,
    config: StackComponentConfig,
    flavor: str,
    type: StackComponentType,
    user: Optional[UUID],
    created: datetime,
    updated: datetime,
    environment: Optional[Dict[str, str]] = None,
    secrets: Optional[List[UUID]] = None,
    labels: Optional[Dict[str, Any]] = None,
    connector_requirements: Optional[ServiceConnectorRequirements] = None,
    connector: Optional[UUID] = None,
    connector_resource_id: Optional[str] = None,
    *args: Any,
    **kwargs: Any,
):
    """Initializes a StackComponent.

    Args:
        name: The name of the component.
        id: The unique ID of the component.
        config: The config of the component.
        flavor: The flavor of the component.
        type: The type of the component.
        user: The ID of the user who created the component.
        created: The creation time of the component.
        updated: The last update time of the component.
        environment: Environment variables to set when running on this
            component.
        secrets: Secrets to set as environment variables when running on
            this component.
        labels: The labels of the component.
        connector_requirements: The requirements for the connector.
        connector: The ID of a connector linked to the component.
        connector_resource_id: The custom resource ID to access through
            the connector.
        *args: Additional positional arguments.
        **kwargs: Additional keyword arguments.

    Raises:
        ValueError: If a secret reference is passed as name.
    """
    if secret_utils.is_secret_reference(name):
        raise ValueError(
            "Passing the `name` attribute of a stack component as a "
            "secret reference is not allowed."
        )

    self.id = id
    self.name = name
    self._config = config
    self.flavor = flavor
    self.type = type
    self.user = user
    self.created = created
    self.updated = updated
    self.labels = labels
    self.environment = environment or {}
    self.secrets = secrets or []
    self.connector_requirements = connector_requirements
    self.connector = connector
    self.connector_resource_id = connector_resource_id
    self._connector_instance: Optional[ServiceConnector] = None

Attributes

config: DatabricksModelDeployerConfig property

Config class for the Databricks Model deployer settings class.

Returns:

Type	Description
DatabricksModelDeployerConfig	The configuration.

validator: Optional[StackValidator] property

Validates the stack.

Returns:

Type	Description
Optional[StackValidator]	A validator that checks that the stack contains a remote artifact
Optional[StackValidator]	store.

Functions

get_model_server_info(service_instance: DatabricksDeploymentService) -> Dict[str, Optional[str]] staticmethod

Return implementation specific information that might be relevant to the user.

Parameters:

Name	Type	Description	Default
service_instance	DatabricksDeploymentService	Instance of a DatabricksDeploymentService	required

Returns:

Type	Description
Dict[str, Optional[str]]	Model server information.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

@staticmethod
def get_model_server_info(  # type: ignore[override]
    service_instance: "DatabricksDeploymentService",
) -> Dict[str, Optional[str]]:
    """Return implementation specific information that might be relevant to the user.

    Args:
        service_instance: Instance of a DatabricksDeploymentService

    Returns:
        Model server information.
    """
    return {
        "PREDICTION_URL": service_instance.get_prediction_url(),
        "HEALTH_CHECK_URL": service_instance.get_healthcheck_url(),
    }

perform_delete_model(service: BaseService, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT, force: bool = False) -> None

Method to delete all configuration of a model server.

Parameters:

Name	Type	Description	Default
service	BaseService	The service to delete.	required
timeout	int	Timeout in seconds to wait for the service to stop.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT
force	bool	If True, force the service to stop.	False

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_delete_model(
    self,
    service: BaseService,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
    force: bool = False,
) -> None:
    """Method to delete all configuration of a model server.

    Args:
        service: The service to delete.
        timeout: Timeout in seconds to wait for the service to stop.
        force: If True, force the service to stop.
    """
    service = cast(DatabricksDeploymentService, service)
    self._clean_up_existing_service(
        existing_service=service, timeout=timeout, force=force
    )

perform_deploy_model(id: UUID, config: ServiceConfig, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT) -> BaseService

Create a new Databricks deployment service or update an existing one.

This should serve the supplied model and deployment configuration.

Parameters:

Name	Type	Description	Default
id	UUID	the UUID of the model to be deployed with Databricks.	required
config	ServiceConfig	the configuration of the model to be deployed with Databricks.	required
timeout	int	the timeout in seconds to wait for the Databricks endpoint to be provisioned and successfully started or updated. If set to 0, the method will return immediately after the Databricks server is provisioned, without waiting for it to fully start.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT

Returns:

Type	Description
BaseService	The ZenML Databricks deployment service object that can be used to
BaseService	interact with the remote Databricks inference endpoint server.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_deploy_model(
    self,
    id: UUID,
    config: ServiceConfig,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
) -> BaseService:
    """Create a new Databricks deployment service or update an existing one.

    This should serve the supplied model and deployment configuration.

    Args:
        id: the UUID of the model to be deployed with Databricks.
        config: the configuration of the model to be deployed with Databricks.
        timeout: the timeout in seconds to wait for the Databricks endpoint
            to be provisioned and successfully started or updated. If set
            to 0, the method will return immediately after the Databricks
            server is provisioned, without waiting for it to fully start.

    Returns:
        The ZenML Databricks deployment service object that can be used to
        interact with the remote Databricks inference endpoint server.
    """
    with track_handler(AnalyticsEvent.MODEL_DEPLOYED) as analytics_handler:
        config = cast(DatabricksDeploymentConfig, config)
        # create a new DatabricksDeploymentService instance
        service = self._create_new_service(
            id=id, timeout=timeout, config=config
        )
        logger.info(
            f"Creating a new Databricks inference endpoint service: {service}"
        )
        # Add telemetry with metadata that gets the stack metadata and
        # differentiates between pure model and custom code deployments
        stack = Client().active_stack
        analytics_handler.metadata = {
            "store_type": Client().zen_store.type.value,
            **stack.component_flavor_metadata,
        }

    return service

perform_start_model(service: BaseService, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT) -> BaseService

Method to start a model server.

Parameters:

Name	Type	Description	Default
service	BaseService	The service to start.	required
timeout	int	Timeout in seconds to wait for the service to start.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT

Returns:

Type	Description
BaseService	The started service.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_start_model(
    self,
    service: BaseService,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
) -> BaseService:
    """Method to start a model server.

    Args:
        service: The service to start.
        timeout: Timeout in seconds to wait for the service to start.

    Returns:
        The started service.
    """
    service.start(timeout=timeout)
    return service

perform_stop_model(service: BaseService, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT, force: bool = False) -> BaseService

Method to stop a model server.

Parameters:

Name	Type	Description	Default
service	BaseService	The service to stop.	required
timeout	int	Timeout in seconds to wait for the service to stop.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT
force	bool	If True, force the service to stop.	False

Returns:

Type	Description
BaseService	The stopped service.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_stop_model(
    self,
    service: BaseService,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
    force: bool = False,
) -> BaseService:
    """Method to stop a model server.

    Args:
        service: The service to stop.
        timeout: Timeout in seconds to wait for the service to stop.
        force: If True, force the service to stop.

    Returns:
        The stopped service.
    """
    service.stop(timeout=timeout, force=force)
    return service

Modules

databricks_model_deployer

Implementation of the Databricks Model Deployer.

Classes

DatabricksModelDeployer(name: str, id: UUID, config: StackComponentConfig, flavor: str, type: StackComponentType, user: Optional[UUID], created: datetime, updated: datetime, environment: Optional[Dict[str, str]] = None, secrets: Optional[List[UUID]] = None, labels: Optional[Dict[str, Any]] = None, connector_requirements: Optional[ServiceConnectorRequirements] = None, connector: Optional[UUID] = None, connector_resource_id: Optional[str] = None, *args: Any, **kwargs: Any)

Bases: BaseModelDeployer

Databricks endpoint model deployer.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self,
    name: str,
    id: UUID,
    config: StackComponentConfig,
    flavor: str,
    type: StackComponentType,
    user: Optional[UUID],
    created: datetime,
    updated: datetime,
    environment: Optional[Dict[str, str]] = None,
    secrets: Optional[List[UUID]] = None,
    labels: Optional[Dict[str, Any]] = None,
    connector_requirements: Optional[ServiceConnectorRequirements] = None,
    connector: Optional[UUID] = None,
    connector_resource_id: Optional[str] = None,
    *args: Any,
    **kwargs: Any,
):
    """Initializes a StackComponent.

    Args:
        name: The name of the component.
        id: The unique ID of the component.
        config: The config of the component.
        flavor: The flavor of the component.
        type: The type of the component.
        user: The ID of the user who created the component.
        created: The creation time of the component.
        updated: The last update time of the component.
        environment: Environment variables to set when running on this
            component.
        secrets: Secrets to set as environment variables when running on
            this component.
        labels: The labels of the component.
        connector_requirements: The requirements for the connector.
        connector: The ID of a connector linked to the component.
        connector_resource_id: The custom resource ID to access through
            the connector.
        *args: Additional positional arguments.
        **kwargs: Additional keyword arguments.

    Raises:
        ValueError: If a secret reference is passed as name.
    """
    if secret_utils.is_secret_reference(name):
        raise ValueError(
            "Passing the `name` attribute of a stack component as a "
            "secret reference is not allowed."
        )

    self.id = id
    self.name = name
    self._config = config
    self.flavor = flavor
    self.type = type
    self.user = user
    self.created = created
    self.updated = updated
    self.labels = labels
    self.environment = environment or {}
    self.secrets = secrets or []
    self.connector_requirements = connector_requirements
    self.connector = connector
    self.connector_resource_id = connector_resource_id
    self._connector_instance: Optional[ServiceConnector] = None

Attributes

config: DatabricksModelDeployerConfig property

Config class for the Databricks Model deployer settings class.

Returns:

Type	Description
DatabricksModelDeployerConfig	The configuration.

validator: Optional[StackValidator] property

Validates the stack.

Returns:

Type	Description
Optional[StackValidator]	A validator that checks that the stack contains a remote artifact
Optional[StackValidator]	store.

Functions

get_model_server_info(service_instance: DatabricksDeploymentService) -> Dict[str, Optional[str]] staticmethod

Return implementation specific information that might be relevant to the user.

Parameters:

Name	Type	Description	Default
service_instance	DatabricksDeploymentService	Instance of a DatabricksDeploymentService	required

Returns:

Type	Description
Dict[str, Optional[str]]	Model server information.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

@staticmethod
def get_model_server_info(  # type: ignore[override]
    service_instance: "DatabricksDeploymentService",
) -> Dict[str, Optional[str]]:
    """Return implementation specific information that might be relevant to the user.

    Args:
        service_instance: Instance of a DatabricksDeploymentService

    Returns:
        Model server information.
    """
    return {
        "PREDICTION_URL": service_instance.get_prediction_url(),
        "HEALTH_CHECK_URL": service_instance.get_healthcheck_url(),
    }

perform_delete_model(service: BaseService, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT, force: bool = False) -> None

Method to delete all configuration of a model server.

Parameters:

Name	Type	Description	Default
service	BaseService	The service to delete.	required
timeout	int	Timeout in seconds to wait for the service to stop.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT
force	bool	If True, force the service to stop.	False

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_delete_model(
    self,
    service: BaseService,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
    force: bool = False,
) -> None:
    """Method to delete all configuration of a model server.

    Args:
        service: The service to delete.
        timeout: Timeout in seconds to wait for the service to stop.
        force: If True, force the service to stop.
    """
    service = cast(DatabricksDeploymentService, service)
    self._clean_up_existing_service(
        existing_service=service, timeout=timeout, force=force
    )

perform_deploy_model(id: UUID, config: ServiceConfig, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT) -> BaseService

Create a new Databricks deployment service or update an existing one.

This should serve the supplied model and deployment configuration.

Parameters:

Name	Type	Description	Default
id	UUID	the UUID of the model to be deployed with Databricks.	required
config	ServiceConfig	the configuration of the model to be deployed with Databricks.	required
timeout	int	the timeout in seconds to wait for the Databricks endpoint to be provisioned and successfully started or updated. If set to 0, the method will return immediately after the Databricks server is provisioned, without waiting for it to fully start.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT

Returns:

Type	Description
BaseService	The ZenML Databricks deployment service object that can be used to
BaseService	interact with the remote Databricks inference endpoint server.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_deploy_model(
    self,
    id: UUID,
    config: ServiceConfig,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
) -> BaseService:
    """Create a new Databricks deployment service or update an existing one.

    This should serve the supplied model and deployment configuration.

    Args:
        id: the UUID of the model to be deployed with Databricks.
        config: the configuration of the model to be deployed with Databricks.
        timeout: the timeout in seconds to wait for the Databricks endpoint
            to be provisioned and successfully started or updated. If set
            to 0, the method will return immediately after the Databricks
            server is provisioned, without waiting for it to fully start.

    Returns:
        The ZenML Databricks deployment service object that can be used to
        interact with the remote Databricks inference endpoint server.
    """
    with track_handler(AnalyticsEvent.MODEL_DEPLOYED) as analytics_handler:
        config = cast(DatabricksDeploymentConfig, config)
        # create a new DatabricksDeploymentService instance
        service = self._create_new_service(
            id=id, timeout=timeout, config=config
        )
        logger.info(
            f"Creating a new Databricks inference endpoint service: {service}"
        )
        # Add telemetry with metadata that gets the stack metadata and
        # differentiates between pure model and custom code deployments
        stack = Client().active_stack
        analytics_handler.metadata = {
            "store_type": Client().zen_store.type.value,
            **stack.component_flavor_metadata,
        }

    return service

perform_start_model(service: BaseService, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT) -> BaseService

Method to start a model server.

Parameters:

Name	Type	Description	Default
service	BaseService	The service to start.	required
timeout	int	Timeout in seconds to wait for the service to start.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT

Returns:

Type	Description
BaseService	The started service.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_start_model(
    self,
    service: BaseService,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
) -> BaseService:
    """Method to start a model server.

    Args:
        service: The service to start.
        timeout: Timeout in seconds to wait for the service to start.

    Returns:
        The started service.
    """
    service.start(timeout=timeout)
    return service

perform_stop_model(service: BaseService, timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT, force: bool = False) -> BaseService

Method to stop a model server.

Parameters:

Name	Type	Description	Default
service	BaseService	The service to stop.	required
timeout	int	Timeout in seconds to wait for the service to stop.	DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT
force	bool	If True, force the service to stop.	False

Returns:

Type	Description
BaseService	The stopped service.

Source code in src/zenml/integrations/databricks/model_deployers/databricks_model_deployer.py

def perform_stop_model(
    self,
    service: BaseService,
    timeout: int = DEFAULT_DEPLOYMENT_START_STOP_TIMEOUT,
    force: bool = False,
) -> BaseService:
    """Method to stop a model server.

    Args:
        service: The service to stop.
        timeout: Timeout in seconds to wait for the service to stop.
        force: If True, force the service to stop.

    Returns:
        The stopped service.
    """
    service.stop(timeout=timeout, force=force)
    return service

Functions

orchestrators

Initialization of the Databricks ZenML orchestrator.

Classes

DatabricksOrchestrator(name: str, id: UUID, config: StackComponentConfig, flavor: str, type: StackComponentType, user: Optional[UUID], created: datetime, updated: datetime, environment: Optional[Dict[str, str]] = None, secrets: Optional[List[UUID]] = None, labels: Optional[Dict[str, Any]] = None, connector_requirements: Optional[ServiceConnectorRequirements] = None, connector: Optional[UUID] = None, connector_resource_id: Optional[str] = None, *args: Any, **kwargs: Any)

Bases: WheeledOrchestrator

Databricks orchestrator.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self,
    name: str,
    id: UUID,
    config: StackComponentConfig,
    flavor: str,
    type: StackComponentType,
    user: Optional[UUID],
    created: datetime,
    updated: datetime,
    environment: Optional[Dict[str, str]] = None,
    secrets: Optional[List[UUID]] = None,
    labels: Optional[Dict[str, Any]] = None,
    connector_requirements: Optional[ServiceConnectorRequirements] = None,
    connector: Optional[UUID] = None,
    connector_resource_id: Optional[str] = None,
    *args: Any,
    **kwargs: Any,
):
    """Initializes a StackComponent.

    Args:
        name: The name of the component.
        id: The unique ID of the component.
        config: The config of the component.
        flavor: The flavor of the component.
        type: The type of the component.
        user: The ID of the user who created the component.
        created: The creation time of the component.
        updated: The last update time of the component.
        environment: Environment variables to set when running on this
            component.
        secrets: Secrets to set as environment variables when running on
            this component.
        labels: The labels of the component.
        connector_requirements: The requirements for the connector.
        connector: The ID of a connector linked to the component.
        connector_resource_id: The custom resource ID to access through
            the connector.
        *args: Additional positional arguments.
        **kwargs: Additional keyword arguments.

    Raises:
        ValueError: If a secret reference is passed as name.
    """
    if secret_utils.is_secret_reference(name):
        raise ValueError(
            "Passing the `name` attribute of a stack component as a "
            "secret reference is not allowed."
        )

    self.id = id
    self.name = name
    self._config = config
    self.flavor = flavor
    self.type = type
    self.user = user
    self.created = created
    self.updated = updated
    self.labels = labels
    self.environment = environment or {}
    self.secrets = secrets or []
    self.connector_requirements = connector_requirements
    self.connector = connector
    self.connector_resource_id = connector_resource_id
    self._connector_instance: Optional[ServiceConnector] = None

Attributes

config: DatabricksOrchestratorConfig property

Returns the DatabricksOrchestratorConfig config.

Returns:

Type	Description
DatabricksOrchestratorConfig	The configuration.

settings_class: Type[DatabricksOrchestratorSettings] property

Settings class for the Databricks orchestrator.

Returns:

Type	Description
Type[DatabricksOrchestratorSettings]	The settings class.

validator: Optional[StackValidator] property

Validates the stack.

Returns:

Type	Description
Optional[StackValidator]	A validator that rejects local stack components.

Functions

get_orchestrator_run_id() -> str

Returns the active orchestrator run id.

Raises:

Type	Description
RuntimeError	If the run id cannot be read from the environment.

Returns:

Type	Description
str	The orchestrator run id.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def get_orchestrator_run_id(self) -> str:
    """Returns the active orchestrator run id.

    Raises:
        RuntimeError: If the run id cannot be read from the environment.

    Returns:
        The orchestrator run id.
    """
    orchestrator_run_id = os.environ.get(
        ENV_ZENML_DATABRICKS_ORCHESTRATOR_RUN_ID
    )
    if orchestrator_run_id is None:
        raise RuntimeError(
            "Unable to read run id from environment variable "
            f"{ENV_ZENML_DATABRICKS_ORCHESTRATOR_RUN_ID}."
        )

    return orchestrator_run_id

get_pipeline_run_metadata(run_id: UUID) -> Dict[str, MetadataType]

Get general component-specific metadata for a pipeline run.

Parameters:

Name	Type	Description	Default
run_id	UUID	The ID of the pipeline run.	required

Returns:

Type	Description
Dict[str, MetadataType]	A dictionary of metadata.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def get_pipeline_run_metadata(
    self, run_id: UUID
) -> Dict[str, "MetadataType"]:
    """Get general component-specific metadata for a pipeline run.

    Args:
        run_id: The ID of the pipeline run.

    Returns:
        A dictionary of metadata.
    """
    run_url = f"{self.config.host}/jobs/{self.get_orchestrator_run_id()}"
    return {
        METADATA_ORCHESTRATOR_URL: Uri(run_url),
    }

setup_credentials() -> None

Set up credentials for the orchestrator.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def setup_credentials(self) -> None:
    """Set up credentials for the orchestrator."""
    connector = self.get_connector()
    assert connector is not None
    connector.configure_local_client()

submit_pipeline(snapshot: PipelineSnapshotResponse, stack: Stack, base_environment: Dict[str, str], step_environments: Dict[str, Dict[str, str]], placeholder_run: Optional[PipelineRunResponse] = None) -> Optional[SubmissionResult]

Submits a pipeline to the orchestrator.

This method should only submit the pipeline and not wait for it to complete. If the orchestrator is configured to wait for the pipeline run to complete, a function that waits for the pipeline run to complete can be passed as part of the submission result.

Parameters:

Name	Type	Description	Default
snapshot	PipelineSnapshotResponse	The pipeline snapshot to submit.	required
stack	Stack	The stack the pipeline will run on.	required
base_environment	Dict[str, str]	Base environment shared by all Databricks tasks.	required
step_environments	Dict[str, Dict[str, str]]	Environment variables to set when executing specific steps.	required
placeholder_run	Optional[PipelineRunResponse]	An optional placeholder run for the snapshot.	None

Raises:

Type	Description
ValueError	If the schedule is not set or if the cron expression is not set.
Exception	If wheel upload or Databricks job submission fails after cleanup.

Returns:

Type	Description
Optional[SubmissionResult]	Optional submission result.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def submit_pipeline(
    self,
    snapshot: "PipelineSnapshotResponse",
    stack: "Stack",
    base_environment: Dict[str, str],
    step_environments: Dict[str, Dict[str, str]],
    placeholder_run: Optional["PipelineRunResponse"] = None,
) -> Optional[SubmissionResult]:
    """Submits a pipeline to the orchestrator.

    This method should only submit the pipeline and not wait for it to
    complete. If the orchestrator is configured to wait for the pipeline run
    to complete, a function that waits for the pipeline run to complete can
    be passed as part of the submission result.

    Args:
        snapshot: The pipeline snapshot to submit.
        stack: The stack the pipeline will run on.
        base_environment: Base environment shared by all Databricks tasks.
        step_environments: Environment variables to set when executing
            specific steps.
        placeholder_run: An optional placeholder run for the snapshot.

    Raises:
        ValueError: If the schedule is not set or if the cron expression
            is not set.
        Exception: If wheel upload or Databricks job submission fails
            after cleanup.

    Returns:
        Optional submission result.
    """
    settings = cast(
        DatabricksOrchestratorSettings, self.get_settings(snapshot)
    )
    if snapshot.schedule:
        if snapshot.schedule.catchup or snapshot.schedule.interval_second:
            logger.warning(
                "Databricks orchestrator only uses schedules with the "
                "`cron_expression` property, with optional `start_time` and/or `end_time`. "
                "All other properties are ignored."
            )
        if snapshot.schedule.cron_expression is None:
            raise ValueError(
                "Property `cron_expression` must be set when passing "
                "schedule to a Databricks orchestrator."
            )
        if (
            snapshot.schedule.cron_expression
            and settings.schedule_timezone is None
        ):
            raise ValueError(
                "Property `schedule_timezone` must be set when passing "
                "`cron_expression` to a Databricks orchestrator. "
                "Databricks orchestrator requires an IANA timezone ID "
                "such as 'America/New_York' or 'UTC' to run the "
                "pipeline on schedule."
            )

    def _construct_databricks_pipeline(
        zenml_project_wheel: str, job_cluster_key: str
    ) -> List[DatabricksTask]:
        """Create one Databricks task for each pipeline step.

        Args:
            zenml_project_wheel: The wheel package containing the ZenML
                project.
            job_cluster_key: The ID of the Databricks job cluster.

        Returns:
            A list of Databricks tasks.
        """
        tasks = []
        for step_name, step in snapshot.step_configurations.items():
            arguments = DatabricksEntrypointConfiguration.get_entrypoint_arguments(
                step_name=step_name,
                snapshot_id=snapshot.id,
                wheel_package=self.package_name,
                databricks_job_id=DATABRICKS_JOB_ID_PARAMETER_REFERENCE,
            )

            # Dependency keys must match the snapshot-prefixed task keys.
            upstream_steps = [
                f"{snapshot.id}_{upstream_step_name}"
                for upstream_step_name in step.spec.upstream_steps
            ]

            task = convert_step_to_task(
                f"{snapshot.id}_{step_name}",
                ZENML_STEP_DEFAULT_ENTRYPOINT_COMMAND,
                arguments,
                collect_requirements(
                    docker_settings=step.config.docker_settings,
                    stack=stack,
                ),
                depends_on=upstream_steps,
                zenml_project_wheel=zenml_project_wheel,
                job_cluster_key=job_cluster_key,
                timeout_seconds=settings.task_timeout_seconds,
                max_retries=settings.max_retries,
                min_retry_interval_millis=settings.min_retry_interval_millis,
                retry_on_timeout=settings.retry_on_timeout,
            )
            tasks.append(task)
        return tasks

    orchestrator_run_name = get_orchestrator_run_name(
        pipeline_name=snapshot.pipeline_configuration.name
    )

    repository_temp_dir = (
        self.copy_repository_to_temp_dir_and_add_setup_py()
    )

    databricks_client = self._get_databricks_client()
    snapshot_name = self.sanitize_name(snapshot.pipeline.name)
    run_namespace = self.sanitize_name(
        str(placeholder_run.id)
        if placeholder_run
        else orchestrator_run_name
    )
    databricks_directory = (
        f"{DATABRICKS_WHEELS_DIRECTORY_PREFIX}/"
        f"{snapshot_name}/{run_namespace}/orchestrator"
    )
    try:
        wheel_path = self.create_wheel(temp_dir=repository_temp_dir)
        try:
            databricks_wheel_path = upload_wheel_to_workspace(
                databricks_client=databricks_client,
                wheel_path=wheel_path,
                databricks_directory=databricks_directory,
            )
        except Exception:
            delete_workspace_directory(
                databricks_client=databricks_client,
                databricks_directory=databricks_directory,
                context="failed wheel upload",
            )
            raise
    finally:
        fileio.rmtree(repository_temp_dir)

    env_vars = base_environment.copy()
    spark_env_vars = settings.spark_env_vars
    if spark_env_vars:
        for key, value in spark_env_vars.items():
            env_vars[key] = value
    env_vars[ENV_ZENML_CUSTOM_SOURCE_ROOT] = (
        DATABRICKS_ZENML_DEFAULT_CUSTOM_REPOSITORY_PATH
    )

    job_cluster_key = self.sanitize_name(str(snapshot.id))
    try:
        self._upload_and_run_pipeline(
            pipeline_name=orchestrator_run_name,
            settings=settings,
            tasks=_construct_databricks_pipeline(
                databricks_wheel_path, job_cluster_key
            ),
            env_vars=env_vars,
            job_cluster_key=job_cluster_key,
            schedule=snapshot.schedule,
        )
    except Exception:
        delete_workspace_directory(
            databricks_client=databricks_client,
            databricks_directory=databricks_directory,
            context="failed job submission",
        )
        raise
    return None

Modules

databricks_orchestrator

Implementation of the Databricks orchestrator.

Classes

DatabricksOrchestrator(name: str, id: UUID, config: StackComponentConfig, flavor: str, type: StackComponentType, user: Optional[UUID], created: datetime, updated: datetime, environment: Optional[Dict[str, str]] = None, secrets: Optional[List[UUID]] = None, labels: Optional[Dict[str, Any]] = None, connector_requirements: Optional[ServiceConnectorRequirements] = None, connector: Optional[UUID] = None, connector_resource_id: Optional[str] = None, *args: Any, **kwargs: Any)

Bases: WheeledOrchestrator

Databricks orchestrator.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self,
    name: str,
    id: UUID,
    config: StackComponentConfig,
    flavor: str,
    type: StackComponentType,
    user: Optional[UUID],
    created: datetime,
    updated: datetime,
    environment: Optional[Dict[str, str]] = None,
    secrets: Optional[List[UUID]] = None,
    labels: Optional[Dict[str, Any]] = None,
    connector_requirements: Optional[ServiceConnectorRequirements] = None,
    connector: Optional[UUID] = None,
    connector_resource_id: Optional[str] = None,
    *args: Any,
    **kwargs: Any,
):
    """Initializes a StackComponent.

    Args:
        name: The name of the component.
        id: The unique ID of the component.
        config: The config of the component.
        flavor: The flavor of the component.
        type: The type of the component.
        user: The ID of the user who created the component.
        created: The creation time of the component.
        updated: The last update time of the component.
        environment: Environment variables to set when running on this
            component.
        secrets: Secrets to set as environment variables when running on
            this component.
        labels: The labels of the component.
        connector_requirements: The requirements for the connector.
        connector: The ID of a connector linked to the component.
        connector_resource_id: The custom resource ID to access through
            the connector.
        *args: Additional positional arguments.
        **kwargs: Additional keyword arguments.

    Raises:
        ValueError: If a secret reference is passed as name.
    """
    if secret_utils.is_secret_reference(name):
        raise ValueError(
            "Passing the `name` attribute of a stack component as a "
            "secret reference is not allowed."
        )

    self.id = id
    self.name = name
    self._config = config
    self.flavor = flavor
    self.type = type
    self.user = user
    self.created = created
    self.updated = updated
    self.labels = labels
    self.environment = environment or {}
    self.secrets = secrets or []
    self.connector_requirements = connector_requirements
    self.connector = connector
    self.connector_resource_id = connector_resource_id
    self._connector_instance: Optional[ServiceConnector] = None

Attributes

config: DatabricksOrchestratorConfig property

Returns the DatabricksOrchestratorConfig config.

Returns:

Type	Description
DatabricksOrchestratorConfig	The configuration.

settings_class: Type[DatabricksOrchestratorSettings] property

Settings class for the Databricks orchestrator.

Returns:

Type	Description
Type[DatabricksOrchestratorSettings]	The settings class.

validator: Optional[StackValidator] property

Validates the stack.

Returns:

Type	Description
Optional[StackValidator]	A validator that rejects local stack components.

Functions

get_orchestrator_run_id() -> str

Returns the active orchestrator run id.

Raises:

Type	Description
RuntimeError	If the run id cannot be read from the environment.

Returns:

Type	Description
str	The orchestrator run id.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def get_orchestrator_run_id(self) -> str:
    """Returns the active orchestrator run id.

    Raises:
        RuntimeError: If the run id cannot be read from the environment.

    Returns:
        The orchestrator run id.
    """
    orchestrator_run_id = os.environ.get(
        ENV_ZENML_DATABRICKS_ORCHESTRATOR_RUN_ID
    )
    if orchestrator_run_id is None:
        raise RuntimeError(
            "Unable to read run id from environment variable "
            f"{ENV_ZENML_DATABRICKS_ORCHESTRATOR_RUN_ID}."
        )

    return orchestrator_run_id

get_pipeline_run_metadata(run_id: UUID) -> Dict[str, MetadataType]

Get general component-specific metadata for a pipeline run.

Parameters:

Name	Type	Description	Default
run_id	UUID	The ID of the pipeline run.	required

Returns:

Type	Description
Dict[str, MetadataType]	A dictionary of metadata.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def get_pipeline_run_metadata(
    self, run_id: UUID
) -> Dict[str, "MetadataType"]:
    """Get general component-specific metadata for a pipeline run.

    Args:
        run_id: The ID of the pipeline run.

    Returns:
        A dictionary of metadata.
    """
    run_url = f"{self.config.host}/jobs/{self.get_orchestrator_run_id()}"
    return {
        METADATA_ORCHESTRATOR_URL: Uri(run_url),
    }

setup_credentials() -> None

Set up credentials for the orchestrator.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def setup_credentials(self) -> None:
    """Set up credentials for the orchestrator."""
    connector = self.get_connector()
    assert connector is not None
    connector.configure_local_client()

submit_pipeline(snapshot: PipelineSnapshotResponse, stack: Stack, base_environment: Dict[str, str], step_environments: Dict[str, Dict[str, str]], placeholder_run: Optional[PipelineRunResponse] = None) -> Optional[SubmissionResult]

Submits a pipeline to the orchestrator.

This method should only submit the pipeline and not wait for it to complete. If the orchestrator is configured to wait for the pipeline run to complete, a function that waits for the pipeline run to complete can be passed as part of the submission result.

Parameters:

Name	Type	Description	Default
snapshot	PipelineSnapshotResponse	The pipeline snapshot to submit.	required
stack	Stack	The stack the pipeline will run on.	required
base_environment	Dict[str, str]	Base environment shared by all Databricks tasks.	required
step_environments	Dict[str, Dict[str, str]]	Environment variables to set when executing specific steps.	required
placeholder_run	Optional[PipelineRunResponse]	An optional placeholder run for the snapshot.	None

Raises:

Type	Description
ValueError	If the schedule is not set or if the cron expression is not set.
Exception	If wheel upload or Databricks job submission fails after cleanup.

Returns:

Type	Description
Optional[SubmissionResult]	Optional submission result.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator.py

def submit_pipeline(
    self,
    snapshot: "PipelineSnapshotResponse",
    stack: "Stack",
    base_environment: Dict[str, str],
    step_environments: Dict[str, Dict[str, str]],
    placeholder_run: Optional["PipelineRunResponse"] = None,
) -> Optional[SubmissionResult]:
    """Submits a pipeline to the orchestrator.

    This method should only submit the pipeline and not wait for it to
    complete. If the orchestrator is configured to wait for the pipeline run
    to complete, a function that waits for the pipeline run to complete can
    be passed as part of the submission result.

    Args:
        snapshot: The pipeline snapshot to submit.
        stack: The stack the pipeline will run on.
        base_environment: Base environment shared by all Databricks tasks.
        step_environments: Environment variables to set when executing
            specific steps.
        placeholder_run: An optional placeholder run for the snapshot.

    Raises:
        ValueError: If the schedule is not set or if the cron expression
            is not set.
        Exception: If wheel upload or Databricks job submission fails
            after cleanup.

    Returns:
        Optional submission result.
    """
    settings = cast(
        DatabricksOrchestratorSettings, self.get_settings(snapshot)
    )
    if snapshot.schedule:
        if snapshot.schedule.catchup or snapshot.schedule.interval_second:
            logger.warning(
                "Databricks orchestrator only uses schedules with the "
                "`cron_expression` property, with optional `start_time` and/or `end_time`. "
                "All other properties are ignored."
            )
        if snapshot.schedule.cron_expression is None:
            raise ValueError(
                "Property `cron_expression` must be set when passing "
                "schedule to a Databricks orchestrator."
            )
        if (
            snapshot.schedule.cron_expression
            and settings.schedule_timezone is None
        ):
            raise ValueError(
                "Property `schedule_timezone` must be set when passing "
                "`cron_expression` to a Databricks orchestrator. "
                "Databricks orchestrator requires an IANA timezone ID "
                "such as 'America/New_York' or 'UTC' to run the "
                "pipeline on schedule."
            )

    def _construct_databricks_pipeline(
        zenml_project_wheel: str, job_cluster_key: str
    ) -> List[DatabricksTask]:
        """Create one Databricks task for each pipeline step.

        Args:
            zenml_project_wheel: The wheel package containing the ZenML
                project.
            job_cluster_key: The ID of the Databricks job cluster.

        Returns:
            A list of Databricks tasks.
        """
        tasks = []
        for step_name, step in snapshot.step_configurations.items():
            arguments = DatabricksEntrypointConfiguration.get_entrypoint_arguments(
                step_name=step_name,
                snapshot_id=snapshot.id,
                wheel_package=self.package_name,
                databricks_job_id=DATABRICKS_JOB_ID_PARAMETER_REFERENCE,
            )

            # Dependency keys must match the snapshot-prefixed task keys.
            upstream_steps = [
                f"{snapshot.id}_{upstream_step_name}"
                for upstream_step_name in step.spec.upstream_steps
            ]

            task = convert_step_to_task(
                f"{snapshot.id}_{step_name}",
                ZENML_STEP_DEFAULT_ENTRYPOINT_COMMAND,
                arguments,
                collect_requirements(
                    docker_settings=step.config.docker_settings,
                    stack=stack,
                ),
                depends_on=upstream_steps,
                zenml_project_wheel=zenml_project_wheel,
                job_cluster_key=job_cluster_key,
                timeout_seconds=settings.task_timeout_seconds,
                max_retries=settings.max_retries,
                min_retry_interval_millis=settings.min_retry_interval_millis,
                retry_on_timeout=settings.retry_on_timeout,
            )
            tasks.append(task)
        return tasks

    orchestrator_run_name = get_orchestrator_run_name(
        pipeline_name=snapshot.pipeline_configuration.name
    )

    repository_temp_dir = (
        self.copy_repository_to_temp_dir_and_add_setup_py()
    )

    databricks_client = self._get_databricks_client()
    snapshot_name = self.sanitize_name(snapshot.pipeline.name)
    run_namespace = self.sanitize_name(
        str(placeholder_run.id)
        if placeholder_run
        else orchestrator_run_name
    )
    databricks_directory = (
        f"{DATABRICKS_WHEELS_DIRECTORY_PREFIX}/"
        f"{snapshot_name}/{run_namespace}/orchestrator"
    )
    try:
        wheel_path = self.create_wheel(temp_dir=repository_temp_dir)
        try:
            databricks_wheel_path = upload_wheel_to_workspace(
                databricks_client=databricks_client,
                wheel_path=wheel_path,
                databricks_directory=databricks_directory,
            )
        except Exception:
            delete_workspace_directory(
                databricks_client=databricks_client,
                databricks_directory=databricks_directory,
                context="failed wheel upload",
            )
            raise
    finally:
        fileio.rmtree(repository_temp_dir)

    env_vars = base_environment.copy()
    spark_env_vars = settings.spark_env_vars
    if spark_env_vars:
        for key, value in spark_env_vars.items():
            env_vars[key] = value
    env_vars[ENV_ZENML_CUSTOM_SOURCE_ROOT] = (
        DATABRICKS_ZENML_DEFAULT_CUSTOM_REPOSITORY_PATH
    )

    job_cluster_key = self.sanitize_name(str(snapshot.id))
    try:
        self._upload_and_run_pipeline(
            pipeline_name=orchestrator_run_name,
            settings=settings,
            tasks=_construct_databricks_pipeline(
                databricks_wheel_path, job_cluster_key
            ),
            env_vars=env_vars,
            job_cluster_key=job_cluster_key,
            schedule=snapshot.schedule,
        )
    except Exception:
        delete_workspace_directory(
            databricks_client=databricks_client,
            databricks_directory=databricks_directory,
            context="failed job submission",
        )
        raise
    return None

Functions Modules

databricks_orchestrator_entrypoint_config

Entrypoint configuration for ZenML Databricks pipeline steps.

Classes

DatabricksEntrypointConfiguration(arguments: List[str])

Bases: StepEntrypointConfiguration

Entrypoint configuration for ZenML Databricks pipeline steps.

Databricks job parameters are limited to 256 characters, so this configuration reconstructs long environment variables from shorter options.

Source code in src/zenml/entrypoints/base_entrypoint_configuration.py

def __init__(self, arguments: List[str]):
    """Initializes the entrypoint configuration.

    Args:
        arguments: Command line arguments to configure this object.
    """
    self.entrypoint_args = self._parse_arguments(arguments)
    self._snapshot: Optional["PipelineSnapshotResponse"] = None

Functions

get_entrypoint_arguments(**kwargs: Any) -> List[str] classmethod

Gets all arguments that the entrypoint command should be called with.

The argument list should be something that argparse.ArgumentParser.parse_args(...) can handle (e.g. ["--some_option", "some_value"] or ["--some_option=some_value"]). It needs to provide values for all options returned by the get_entrypoint_options() method of this class.

Parameters:

Name	Type	Description	Default
**kwargs	Any	Kwargs, must include the step name.	{}

Returns:

Type	Description
List[str]	The superclass arguments as well as arguments for the wheel package.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator_entrypoint_config.py

@classmethod
def get_entrypoint_arguments(
    cls,
    **kwargs: Any,
) -> List[str]:
    """Gets all arguments that the entrypoint command should be called with.

    The argument list should be something that
    `argparse.ArgumentParser.parse_args(...)` can handle (e.g.
    `["--some_option", "some_value"]` or `["--some_option=some_value"]`).
    It needs to provide values for all options returned by the
    `get_entrypoint_options()` method of this class.

    Args:
        **kwargs: Kwargs, must include the step name.

    Returns:
        The superclass arguments as well as arguments for the wheel package.
    """
    return super().get_entrypoint_arguments(**kwargs) + [
        f"--{WHEEL_PACKAGE_OPTION}",
        kwargs[WHEEL_PACKAGE_OPTION],
        f"--{DATABRICKS_JOB_ID_OPTION}",
        kwargs[DATABRICKS_JOB_ID_OPTION],
    ]

get_entrypoint_options() -> Dict[str, bool] classmethod

Gets all options required for running with this configuration.

Returns:

Type	Description
Dict[str, bool]	The superclass options as well as an option for the wheel package.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator_entrypoint_config.py

@classmethod
def get_entrypoint_options(cls) -> Dict[str, bool]:
    """Gets all options required for running with this configuration.

    Returns:
        The superclass options as well as an option for the wheel package.
    """
    return super().get_entrypoint_options() | {
        WHEEL_PACKAGE_OPTION: True,
        DATABRICKS_JOB_ID_OPTION: True,
    }

run() -> None

Runs the step.

Source code in src/zenml/integrations/databricks/orchestrators/databricks_orchestrator_entrypoint_config.py

def run(self) -> None:
    """Runs the step."""
    wheel_package = self.entrypoint_args[WHEEL_PACKAGE_OPTION]
    configure_databricks_wheel_environment(wheel_package)

    databricks_job_id = self.entrypoint_args[DATABRICKS_JOB_ID_OPTION]
    os.environ[ENV_ZENML_DATABRICKS_ORCHESTRATOR_RUN_ID] = (
        databricks_job_id
    )

    super().run()

Functions

services

Initialization of the Databricks Service.

Classes

Modules

databricks_deployment

Implementation of the Databricks Deployment service.

Classes

DatabricksDeploymentConfig(**data: Any)

Bases: DatabricksBaseConfig, ServiceConfig

Databricks service configurations.

Source code in src/zenml/services/service.py

def __init__(self, **data: Any):
    """Initialize the service configuration.

    Args:
        **data: keyword arguments.

    Raises:
        ValueError: if neither 'name' nor 'model_name' is set.
    """
    super().__init__(**data)
    if self.name or self.model_name:
        self.service_name = data.get(
            "service_name",
            f"{ZENM_ENDPOINT_PREFIX}{self.name or self.model_name}",
        )
    else:
        raise ValueError("Either 'name' or 'model_name' must be set.")

Functions

get_databricks_deployment_labels() -> Dict[str, str]

Generate labels for the Databricks deployment from the service configuration.

These labels are attached to the Databricks deployment resource and may be used as label selectors in lookup operations.

Returns:

Type	Description
Dict[str, str]	The labels for the Databricks deployment.

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def get_databricks_deployment_labels(self) -> Dict[str, str]:
    """Generate labels for the Databricks deployment from the service configuration.

    These labels are attached to the Databricks deployment resource
    and may be used as label selectors in lookup operations.

    Returns:
        The labels for the Databricks deployment.
    """
    labels = {}
    if self.pipeline_name:
        labels["zenml_pipeline_name"] = self.pipeline_name
    if self.pipeline_step_name:
        labels["zenml_pipeline_step_name"] = self.pipeline_step_name
    if self.model_name:
        labels["zenml_model_name"] = self.model_name
    if self.model_uri:
        labels["zenml_model_uri"] = self.model_uri
    sanitize_labels(labels)
    return labels

DatabricksDeploymentService(config: DatabricksDeploymentConfig, **attrs: Any)

Bases: BaseDeploymentService

Databricks model deployment service.

Attributes:

Name	Type	Description
SERVICE_TYPE		a service type descriptor with information describing the Databricks deployment service class
config	DatabricksDeploymentConfig	service configuration

Initialize the Databricks deployment service.

Parameters:

Name	Type	Description	Default
config	DatabricksDeploymentConfig	service configuration	required
attrs	Any	additional attributes to set on the service	{}

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def __init__(self, config: DatabricksDeploymentConfig, **attrs: Any):
    """Initialize the Databricks deployment service.

    Args:
        config: service configuration
        attrs: additional attributes to set on the service
    """
    super().__init__(config=config, **attrs)

Attributes

databricks_client: DatabricksClient property

Get the deployed Databricks inference endpoint.

Returns:

Type	Description
WorkspaceClient	databricks inference endpoint.

databricks_endpoint: ServingEndpointDetailed property

Get the deployed Hugging Face inference endpoint.

Returns:

Type	Description
ServingEndpointDetailed	Databricks inference endpoint.

prediction_url: Optional[str] property

The prediction URI exposed by the prediction service.

Returns:

Type	Description
Optional[str]	The prediction URI exposed by the prediction service, or None if
Optional[str]	the service is not yet ready.

Functions

check_status() -> Tuple[ServiceState, str]

Check the current operational state of the Databricks deployment.

Returns:

Type	Description
ServiceState	The operational state of the Databricks deployment and a message
str	providing additional information about that state (e.g. a
Tuple[ServiceState, str]	description of the error, if one is encountered).

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def check_status(self) -> Tuple[ServiceState, str]:
    """Check the current operational state of the Databricks deployment.

    Returns:
        The operational state of the Databricks deployment and a message
        providing additional information about that state (e.g. a
        description of the error, if one is encountered).
    """
    try:
        status = self.databricks_endpoint.state or None
        if (
            status
            and status.ready
            and status.ready == EndpointStateReady.READY
        ):
            return (ServiceState.ACTIVE, "")
        elif (
            status
            and status.config_update
            and status.config_update
            == EndpointStateConfigUpdate.UPDATE_FAILED
        ):
            return (
                ServiceState.ERROR,
                "Databricks Inference Endpoint deployment update failed",
            )
        elif (
            status
            and status.config_update
            and status.config_update
            == EndpointStateConfigUpdate.IN_PROGRESS
        ):
            return (ServiceState.PENDING_STARTUP, "")
        return (ServiceState.PENDING_STARTUP, "")
    except Exception as e:
        return (
            ServiceState.INACTIVE,
            f"Databricks Inference Endpoint deployment is inactive or not found: {e}",
        )

deprovision(force: bool = False) -> None

Deprovision the remote Databricks deployment instance.

Parameters:

Name	Type	Description	Default
force	bool	if True, the remote deployment instance will be forcefully deprovisioned.	False

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def deprovision(self, force: bool = False) -> None:
    """Deprovision the remote Databricks deployment instance.

    Args:
        force: if True, the remote deployment instance will be
            forcefully deprovisioned.
    """
    try:
        self.databricks_client.serving_endpoints.delete(
            name=self._generate_an_endpoint_name()
        )
    except Exception:
        logger.error(
            "Databricks Inference Endpoint is deleted or cannot be found."
        )

get_client_id_and_secret() -> Tuple[str, str, str]

Get the Databricks client id and secret.

Raises:

Type	Description
ValueError	If client id and secret are not found.

Returns:

Type	Description
Tuple[str, str, str]	Databricks client id and secret.

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def get_client_id_and_secret(self) -> Tuple[str, str, str]:
    """Get the Databricks client id and secret.

    Raises:
        ValueError: If client id and secret are not found.

    Returns:
        Databricks client id and secret.
    """
    client = Client()
    client_id = None
    client_secret = None
    host = None
    from zenml.integrations.databricks.model_deployers.databricks_model_deployer import (
        DatabricksModelDeployer,
    )

    model_deployer = client.active_stack.model_deployer
    if not isinstance(model_deployer, DatabricksModelDeployer):
        raise ValueError(
            "DatabricksModelDeployer is not active in the stack."
        )
    host = model_deployer.config.host
    self.config.host = host
    if model_deployer.config.secret_name:
        secret = client.get_secret(model_deployer.config.secret_name)
        client_id = secret.secret_values["client_id"]
        client_secret = secret.secret_values["client_secret"]

    else:
        client_id = model_deployer.config.client_id
        client_secret = model_deployer.config.client_secret
    if not client_id:
        raise ValueError("Client id not found.")
    if not client_secret:
        raise ValueError("Client secret not found.")
    if not host:
        raise ValueError("Host not found.")
    return host, client_id, client_secret

get_logs(follow: bool = False, tail: Optional[int] = None) -> Generator[str, bool, None]

Retrieve the service logs.

Parameters:

Name	Type	Description	Default
follow	bool	if True, the logs will be streamed as they are written	False
tail	Optional[int]	only retrieve the last NUM lines of log output.	None

Yields:

Type	Description
str	A generator that can be accessed to get the service logs.

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def get_logs(
    self, follow: bool = False, tail: Optional[int] = None
) -> Generator[str, bool, None]:
    """Retrieve the service logs.

    Args:
        follow: if True, the logs will be streamed as they are written
        tail: only retrieve the last NUM lines of log output.

    Yields:
        A generator that can be accessed to get the service logs.
    """
    logger.info(
        "Databricks Endpoints provides access to the logs of your Endpoints through the UI in the `Logs` tab of your Endpoint"
    )

    def log_generator() -> Generator[str, bool, None]:
        last_log_count = 0
        while True:
            logs = self.databricks_client.serving_endpoints.logs(
                name=self._generate_an_endpoint_name(),
                served_model_name=self.config.model_name,
            )

            log_lines = logs.logs.split("\n")

            # Apply tail if specified and it's the first iteration
            if tail is not None and last_log_count == 0:
                log_lines = log_lines[-tail:]

            # Yield only new lines
            for line in log_lines[last_log_count:]:
                yield line

            last_log_count = len(log_lines)

            if not follow:
                break

            # Add a small delay to avoid excessive API calls
            time.sleep(1)

    yield from log_generator()

predict(request: Union[NDArray[Any], pd.DataFrame]) -> NDArray[Any]

Make a prediction using the service.

Parameters:

Name	Type	Description	Default
request	Union[NDArray[Any], DataFrame]	The input data for the prediction.	required

Returns:

Type	Description
NDArray[Any]	The prediction result.

Raises:

Type	Description
Exception	if the service is not running
ValueError	if the endpoint secret name is not provided.

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def predict(
    self, request: Union["NDArray[Any]", pd.DataFrame]
) -> "NDArray[Any]":
    """Make a prediction using the service.

    Args:
        request: The input data for the prediction.

    Returns:
        The prediction result.

    Raises:
        Exception: if the service is not running
        ValueError: if the endpoint secret name is not provided.
    """
    if not self.is_running:
        raise Exception(
            "Databricks endpoint inference service is not running. "
            "Please start the service before making predictions."
        )
    if self.prediction_url is not None:
        if not self.config.endpoint_secret_name:
            raise ValueError(
                "No endpoint secret name is provided for prediction."
            )
        databricks_token = Client().get_secret(
            self.config.endpoint_secret_name
        )
        if not databricks_token.secret_values["token"]:
            raise ValueError("No databricks token found.")
        headers = {
            "Authorization": f"Bearer {databricks_token.secret_values['token']}",
            "Content-Type": "application/json",
        }
        if isinstance(request, pd.DataFrame):
            response = requests.post(  # nosec
                self.prediction_url,
                json={"instances": request.to_dict("records")},
                headers=headers,
            )
        else:
            response = requests.post(  # nosec
                self.prediction_url,
                json={"instances": request.tolist()},
                headers=headers,
            )
    else:
        raise ValueError("No endpoint known for prediction.")
    response.raise_for_status()

    return np.array(response.json()["predictions"])

provision() -> None

Provision or update remote Databricks deployment instance.

Source code in src/zenml/integrations/databricks/services/databricks_deployment.py

def provision(self) -> None:
    """Provision or update remote Databricks deployment instance."""
    from databricks.sdk.service.serving import (
        ServedModelInputWorkloadSize,
        ServedModelInputWorkloadType,
    )

    tags = []
    for key, value in self._get_databricks_deployment_labels().items():
        tags.append(EndpointTag(key=key, value=value))
    # Attempt to create and wait for the inference endpoint
    served_model = ServedModelInput(
        model_name=self.config.model_name,
        model_version=self.config.model_version,
        scale_to_zero_enabled=self.config.scale_to_zero_enabled,
        workload_type=ServedModelInputWorkloadType(
            self.config.workload_type
        ),
        workload_size=ServedModelInputWorkloadSize(
            self.config.workload_size
        ),
    )

    databricks_endpoint = (
        self.databricks_client.serving_endpoints.create_and_wait(
            name=self._generate_an_endpoint_name(),
            config=EndpointCoreConfigInput(
                served_models=[served_model],
            ),
            tags=tags,
        )
    )
    # Check if the endpoint URL is available after provisioning
    if databricks_endpoint.endpoint_url:
        logger.info(
            f"Databricks inference endpoint successfully deployed and available. Endpoint URL: {databricks_endpoint.endpoint_url}"
        )
    else:
        logger.error(
            "Failed to start Databricks inference endpoint service: No URL available, please check the Databricks console for more details."
        )

DatabricksServiceStatus

Bases: ServiceStatus

Databricks service status.

Functions

step_operators

Initialization of the Databricks step operators.

Classes

DatabricksStepOperator(name: str, id: UUID, config: StackComponentConfig, flavor: str, type: StackComponentType, user: Optional[UUID], created: datetime, updated: datetime, environment: Optional[Dict[str, str]] = None, secrets: Optional[List[UUID]] = None, labels: Optional[Dict[str, Any]] = None, connector_requirements: Optional[ServiceConnectorRequirements] = None, connector: Optional[UUID] = None, connector_resource_id: Optional[str] = None, *args: Any, **kwargs: Any)

Bases: BaseStepOperator

Step operator to run individual steps on Databricks.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self,
    name: str,
    id: UUID,
    config: StackComponentConfig,
    flavor: str,
    type: StackComponentType,
    user: Optional[UUID],
    created: datetime,
    updated: datetime,
    environment: Optional[Dict[str, str]] = None,
    secrets: Optional[List[UUID]] = None,
    labels: Optional[Dict[str, Any]] = None,
    connector_requirements: Optional[ServiceConnectorRequirements] = None,
    connector: Optional[UUID] = None,
    connector_resource_id: Optional[str] = None,
    *args: Any,
    **kwargs: Any,
):
    """Initializes a StackComponent.

    Args:
        name: The name of the component.
        id: The unique ID of the component.
        config: The config of the component.
        flavor: The flavor of the component.
        type: The type of the component.
        user: The ID of the user who created the component.
        created: The creation time of the component.
        updated: The last update time of the component.
        environment: Environment variables to set when running on this
            component.
        secrets: Secrets to set as environment variables when running on
            this component.
        labels: The labels of the component.
        connector_requirements: The requirements for the connector.
        connector: The ID of a connector linked to the component.
        connector_resource_id: The custom resource ID to access through
            the connector.
        *args: Additional positional arguments.
        **kwargs: Additional keyword arguments.

    Raises:
        ValueError: If a secret reference is passed as name.
    """
    if secret_utils.is_secret_reference(name):
        raise ValueError(
            "Passing the `name` attribute of a stack component as a "
            "secret reference is not allowed."
        )

    self.id = id
    self.name = name
    self._config = config
    self.flavor = flavor
    self.type = type
    self.user = user
    self.created = created
    self.updated = updated
    self.labels = labels
    self.environment = environment or {}
    self.secrets = secrets or []
    self.connector_requirements = connector_requirements
    self.connector = connector
    self.connector_resource_id = connector_resource_id
    self._connector_instance: Optional[ServiceConnector] = None

Attributes

client: DatabricksClient property

Get or create the Databricks client.

Returns:

Type	Description
WorkspaceClient	The Databricks client instance.

config: DatabricksStepOperatorConfig property

Returns the step operator config.

Returns:

Type	Description
DatabricksStepOperatorConfig	The configuration.

entrypoint_config_class: Type[DatabricksStepOperatorEntrypointConfiguration] property

Returns the entrypoint configuration class for Databricks steps.

Returns:

Type	Description
Type[DatabricksStepOperatorEntrypointConfiguration]	The Databricks step operator entrypoint configuration class.

settings_class: Optional[Type[BaseSettings]] property

Settings class for the Databricks step operator.

Returns:

Type	Description
Optional[Type[BaseSettings]]	The settings class.

validator: Optional[StackValidator] property

Validates the stack for Databricks step operator execution.

Returns:

Type	Description
Optional[StackValidator]	A validator that checks for a remote artifact store.

Functions

cancel(step_run: StepRunResponse) -> None

Cancel a Databricks step run.

Parameters:

Name	Type	Description	Default
step_run	StepRunResponse	The step run to cancel.	required

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator.py

def cancel(self, step_run: "StepRunResponse") -> None:
    """Cancel a Databricks step run.

    Args:
        step_run: The step run to cancel.
    """
    run_id = self._get_run_id(step_run)

    try:
        self.client.jobs.cancel_run(run_id=run_id)
    except NotFound:
        logger.warning(
            "Databricks run `%s` for step run `%s` was not found during "
            "cancellation.",
            run_id,
            step_run.id,
        )

get_status(step_run: StepRunResponse) -> ExecutionStatus

Get the status of a Databricks step run.

Parameters:

Name	Type	Description	Default
step_run	StepRunResponse	The step run to get the status of.	required

Returns:

Type	Description
ExecutionStatus	The ZenML step execution status.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator.py

def get_status(self, step_run: "StepRunResponse") -> ExecutionStatus:
    """Get the status of a Databricks step run.

    Args:
        step_run: The step run to get the status of.

    Returns:
        The ZenML step execution status.
    """
    run_id = self._get_run_id(step_run)

    try:
        run = self.client.jobs.get_run(run_id=run_id)
    except NotFound:
        logger.warning(
            "Databricks run `%s` for step run `%s` was not found.",
            run_id,
            step_run.id,
        )
        return ExecutionStatus.FAILED

    status = map_databricks_run_to_execution_status(run)
    if status == ExecutionStatus.COMPLETED:
        self._cleanup_uploaded_wheel_directory(step_run)
    return status

submit(info: StepRunInfo, entrypoint_command: List[str], environment: Dict[str, str]) -> None

Submit a step run to Databricks.

Parameters:

Name	Type	Description	Default
info	StepRunInfo	Information about the step run.	required
entrypoint_command	List[str]	Command that executes the step.	required
environment	Dict[str, str]	Environment variables to set in the step operator environment.	required

Raises:

Type	Description
Exception	If wheel upload or Databricks run submission fails after cleanup.
RuntimeError	If the stack or Databricks run ID is missing.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator.py

def submit(
    self,
    info: "StepRunInfo",
    entrypoint_command: List[str],
    environment: Dict[str, str],
) -> None:
    """Submit a step run to Databricks.

    Args:
        info: Information about the step run.
        entrypoint_command: Command that executes the step.
        environment: Environment variables to set in the step operator
            environment.

    Raises:
        Exception: If wheel upload or Databricks run submission fails
            after cleanup.
        RuntimeError: If the stack or Databricks run ID is missing.
    """
    settings = cast(
        DatabricksStepOperatorSettings, self.get_settings(info)
    )
    stack_model = info.snapshot.stack
    if stack_model is None:
        raise RuntimeError(
            "Missing stack for Databricks step operator submission."
        )

    info.force_write_logs()

    wheel_source = get_databricks_wheel_source()
    if wheel_source:
        source_root, package_name = wheel_source
    else:
        source_root = None
        package_name = get_wheel_package_name()
    repository_temp_dir = prepare_repository_copy_for_wheel(
        package_name=package_name,
        package_version=__version__,
        source_root=source_root,
    )
    databricks_directory = (
        f"{DATABRICKS_WHEELS_DIRECTORY_PREFIX}/"
        f"{sanitize_name(info.pipeline.name)}/"
        f"{sanitize_name(str(info.run_id))}/steps/{info.step_run_id}"
    )
    try:
        wheel_path = create_wheel(repository_temp_dir)
        try:
            databricks_wheel_path = upload_wheel_to_workspace(
                databricks_client=self.client,
                wheel_path=wheel_path,
                databricks_directory=databricks_directory,
            )
        except Exception:
            delete_workspace_directory(
                databricks_client=self.client,
                databricks_directory=databricks_directory,
                context="failed wheel upload",
            )
            raise

        try:
            environment = environment.copy()
            if settings.spark_env_vars:
                environment.update(settings.spark_env_vars)
            environment[ENV_ZENML_CUSTOM_SOURCE_ROOT] = (
                DATABRICKS_ZENML_DEFAULT_CUSTOM_REPOSITORY_PATH
            )

            entrypoint_arguments = entrypoint_command[1:] + [
                f"--{WHEEL_PACKAGE_OPTION}",
                package_name,
            ]
            task = convert_step_to_submit_task(
                task_name=str(info.step_run_id),
                command=entrypoint_command[0],
                arguments=entrypoint_arguments,
                libraries=collect_requirements(
                    docker_settings=info.config.docker_settings,
                    stack=Stack.from_model(stack_model),
                ),
                zenml_project_wheel=databricks_wheel_path,
                new_cluster=build_databricks_cluster_spec(
                    databricks_client=self.client,
                    settings=settings,
                    env_vars=environment,
                ),
                timeout_seconds=settings.task_timeout_seconds,
            )
            submitted_run = self.client.jobs.submit(
                access_control_list=build_access_control_list(settings),
                run_name=f"{info.run_name}-{info.pipeline_step_name}",
                tasks=[task],
                timeout_seconds=settings.timeout_seconds,
            )
        except Exception:
            delete_workspace_directory(
                databricks_client=self.client,
                databricks_directory=databricks_directory,
                context="failed run submission",
            )
            raise
    finally:
        fileio.rmtree(repository_temp_dir)

    submit_response = cast(SubmitRunResponse, submitted_run.response)
    run_id = submit_response.run_id
    if run_id is None:
        delete_workspace_directory(
            databricks_client=self.client,
            databricks_directory=databricks_directory,
            context="run submission without run ID",
        )
        raise RuntimeError(
            "Databricks did not return a run ID for the submitted step."
        )

    metadata: Dict[str, MetadataType] = {
        DATABRICKS_STEP_RUN_ID_METADATA_KEY: str(run_id),
        DATABRICKS_STEP_WHEEL_DIRECTORY_METADATA_KEY: (
            databricks_directory
        ),
    }
    publish_step_run_metadata(info.step_run_id, {self.id: metadata})
    info.step_run.run_metadata.update(metadata)
    try:
        run = self.client.jobs.get_run(run_id=run_id)
    except NotFound:
        logger.warning(
            "Submitted Databricks run '%s' for step '%s', but the run "
            "was not found while retrieving additional metadata.",
            run_id,
            info.pipeline_step_name,
        )
    else:
        extra_metadata: Dict[str, MetadataType] = {}
        if run.job_id is not None:
            extra_metadata[DATABRICKS_STEP_JOB_ID_METADATA_KEY] = str(
                run.job_id
            )
        if run.run_page_url:
            extra_metadata[DATABRICKS_STEP_RUN_URL_METADATA_KEY] = (
                run.run_page_url
            )
        if extra_metadata:
            publish_step_run_metadata(
                info.step_run_id, {self.id: extra_metadata}
            )
            info.step_run.run_metadata.update(extra_metadata)

    logger.info(
        "Submitted step '%s' to Databricks as run '%s'.",
        info.pipeline_step_name,
        run_id,
    )

Modules

databricks_step_operator

Databricks step operator implementation.

Classes

DatabricksStepOperator(name: str, id: UUID, config: StackComponentConfig, flavor: str, type: StackComponentType, user: Optional[UUID], created: datetime, updated: datetime, environment: Optional[Dict[str, str]] = None, secrets: Optional[List[UUID]] = None, labels: Optional[Dict[str, Any]] = None, connector_requirements: Optional[ServiceConnectorRequirements] = None, connector: Optional[UUID] = None, connector_resource_id: Optional[str] = None, *args: Any, **kwargs: Any)

Bases: BaseStepOperator

Step operator to run individual steps on Databricks.

Source code in src/zenml/stack/stack_component.py

def __init__(
    self,
    name: str,
    id: UUID,
    config: StackComponentConfig,
    flavor: str,
    type: StackComponentType,
    user: Optional[UUID],
    created: datetime,
    updated: datetime,
    environment: Optional[Dict[str, str]] = None,
    secrets: Optional[List[UUID]] = None,
    labels: Optional[Dict[str, Any]] = None,
    connector_requirements: Optional[ServiceConnectorRequirements] = None,
    connector: Optional[UUID] = None,
    connector_resource_id: Optional[str] = None,
    *args: Any,
    **kwargs: Any,
):
    """Initializes a StackComponent.

    Args:
        name: The name of the component.
        id: The unique ID of the component.
        config: The config of the component.
        flavor: The flavor of the component.
        type: The type of the component.
        user: The ID of the user who created the component.
        created: The creation time of the component.
        updated: The last update time of the component.
        environment: Environment variables to set when running on this
            component.
        secrets: Secrets to set as environment variables when running on
            this component.
        labels: The labels of the component.
        connector_requirements: The requirements for the connector.
        connector: The ID of a connector linked to the component.
        connector_resource_id: The custom resource ID to access through
            the connector.
        *args: Additional positional arguments.
        **kwargs: Additional keyword arguments.

    Raises:
        ValueError: If a secret reference is passed as name.
    """
    if secret_utils.is_secret_reference(name):
        raise ValueError(
            "Passing the `name` attribute of a stack component as a "
            "secret reference is not allowed."
        )

    self.id = id
    self.name = name
    self._config = config
    self.flavor = flavor
    self.type = type
    self.user = user
    self.created = created
    self.updated = updated
    self.labels = labels
    self.environment = environment or {}
    self.secrets = secrets or []
    self.connector_requirements = connector_requirements
    self.connector = connector
    self.connector_resource_id = connector_resource_id
    self._connector_instance: Optional[ServiceConnector] = None

Attributes

client: DatabricksClient property

Get or create the Databricks client.

Returns:

Type	Description
WorkspaceClient	The Databricks client instance.

config: DatabricksStepOperatorConfig property

Returns the step operator config.

Returns:

Type	Description
DatabricksStepOperatorConfig	The configuration.

entrypoint_config_class: Type[DatabricksStepOperatorEntrypointConfiguration] property

Returns the entrypoint configuration class for Databricks steps.

Returns:

Type	Description
Type[DatabricksStepOperatorEntrypointConfiguration]	The Databricks step operator entrypoint configuration class.

settings_class: Optional[Type[BaseSettings]] property

Settings class for the Databricks step operator.

Returns:

Type	Description
Optional[Type[BaseSettings]]	The settings class.

validator: Optional[StackValidator] property

Validates the stack for Databricks step operator execution.

Returns:

Type	Description
Optional[StackValidator]	A validator that checks for a remote artifact store.

Functions

cancel(step_run: StepRunResponse) -> None

Cancel a Databricks step run.

Parameters:

Name	Type	Description	Default
step_run	StepRunResponse	The step run to cancel.	required

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator.py

def cancel(self, step_run: "StepRunResponse") -> None:
    """Cancel a Databricks step run.

    Args:
        step_run: The step run to cancel.
    """
    run_id = self._get_run_id(step_run)

    try:
        self.client.jobs.cancel_run(run_id=run_id)
    except NotFound:
        logger.warning(
            "Databricks run `%s` for step run `%s` was not found during "
            "cancellation.",
            run_id,
            step_run.id,
        )

get_status(step_run: StepRunResponse) -> ExecutionStatus

Get the status of a Databricks step run.

Parameters:

Name	Type	Description	Default
step_run	StepRunResponse	The step run to get the status of.	required

Returns:

Type	Description
ExecutionStatus	The ZenML step execution status.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator.py

def get_status(self, step_run: "StepRunResponse") -> ExecutionStatus:
    """Get the status of a Databricks step run.

    Args:
        step_run: The step run to get the status of.

    Returns:
        The ZenML step execution status.
    """
    run_id = self._get_run_id(step_run)

    try:
        run = self.client.jobs.get_run(run_id=run_id)
    except NotFound:
        logger.warning(
            "Databricks run `%s` for step run `%s` was not found.",
            run_id,
            step_run.id,
        )
        return ExecutionStatus.FAILED

    status = map_databricks_run_to_execution_status(run)
    if status == ExecutionStatus.COMPLETED:
        self._cleanup_uploaded_wheel_directory(step_run)
    return status

submit(info: StepRunInfo, entrypoint_command: List[str], environment: Dict[str, str]) -> None

Submit a step run to Databricks.

Parameters:

Name	Type	Description	Default
info	StepRunInfo	Information about the step run.	required
entrypoint_command	List[str]	Command that executes the step.	required
environment	Dict[str, str]	Environment variables to set in the step operator environment.	required

Raises:

Type	Description
Exception	If wheel upload or Databricks run submission fails after cleanup.
RuntimeError	If the stack or Databricks run ID is missing.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator.py

def submit(
    self,
    info: "StepRunInfo",
    entrypoint_command: List[str],
    environment: Dict[str, str],
) -> None:
    """Submit a step run to Databricks.

    Args:
        info: Information about the step run.
        entrypoint_command: Command that executes the step.
        environment: Environment variables to set in the step operator
            environment.

    Raises:
        Exception: If wheel upload or Databricks run submission fails
            after cleanup.
        RuntimeError: If the stack or Databricks run ID is missing.
    """
    settings = cast(
        DatabricksStepOperatorSettings, self.get_settings(info)
    )
    stack_model = info.snapshot.stack
    if stack_model is None:
        raise RuntimeError(
            "Missing stack for Databricks step operator submission."
        )

    info.force_write_logs()

    wheel_source = get_databricks_wheel_source()
    if wheel_source:
        source_root, package_name = wheel_source
    else:
        source_root = None
        package_name = get_wheel_package_name()
    repository_temp_dir = prepare_repository_copy_for_wheel(
        package_name=package_name,
        package_version=__version__,
        source_root=source_root,
    )
    databricks_directory = (
        f"{DATABRICKS_WHEELS_DIRECTORY_PREFIX}/"
        f"{sanitize_name(info.pipeline.name)}/"
        f"{sanitize_name(str(info.run_id))}/steps/{info.step_run_id}"
    )
    try:
        wheel_path = create_wheel(repository_temp_dir)
        try:
            databricks_wheel_path = upload_wheel_to_workspace(
                databricks_client=self.client,
                wheel_path=wheel_path,
                databricks_directory=databricks_directory,
            )
        except Exception:
            delete_workspace_directory(
                databricks_client=self.client,
                databricks_directory=databricks_directory,
                context="failed wheel upload",
            )
            raise

        try:
            environment = environment.copy()
            if settings.spark_env_vars:
                environment.update(settings.spark_env_vars)
            environment[ENV_ZENML_CUSTOM_SOURCE_ROOT] = (
                DATABRICKS_ZENML_DEFAULT_CUSTOM_REPOSITORY_PATH
            )

            entrypoint_arguments = entrypoint_command[1:] + [
                f"--{WHEEL_PACKAGE_OPTION}",
                package_name,
            ]
            task = convert_step_to_submit_task(
                task_name=str(info.step_run_id),
                command=entrypoint_command[0],
                arguments=entrypoint_arguments,
                libraries=collect_requirements(
                    docker_settings=info.config.docker_settings,
                    stack=Stack.from_model(stack_model),
                ),
                zenml_project_wheel=databricks_wheel_path,
                new_cluster=build_databricks_cluster_spec(
                    databricks_client=self.client,
                    settings=settings,
                    env_vars=environment,
                ),
                timeout_seconds=settings.task_timeout_seconds,
            )
            submitted_run = self.client.jobs.submit(
                access_control_list=build_access_control_list(settings),
                run_name=f"{info.run_name}-{info.pipeline_step_name}",
                tasks=[task],
                timeout_seconds=settings.timeout_seconds,
            )
        except Exception:
            delete_workspace_directory(
                databricks_client=self.client,
                databricks_directory=databricks_directory,
                context="failed run submission",
            )
            raise
    finally:
        fileio.rmtree(repository_temp_dir)

    submit_response = cast(SubmitRunResponse, submitted_run.response)
    run_id = submit_response.run_id
    if run_id is None:
        delete_workspace_directory(
            databricks_client=self.client,
            databricks_directory=databricks_directory,
            context="run submission without run ID",
        )
        raise RuntimeError(
            "Databricks did not return a run ID for the submitted step."
        )

    metadata: Dict[str, MetadataType] = {
        DATABRICKS_STEP_RUN_ID_METADATA_KEY: str(run_id),
        DATABRICKS_STEP_WHEEL_DIRECTORY_METADATA_KEY: (
            databricks_directory
        ),
    }
    publish_step_run_metadata(info.step_run_id, {self.id: metadata})
    info.step_run.run_metadata.update(metadata)
    try:
        run = self.client.jobs.get_run(run_id=run_id)
    except NotFound:
        logger.warning(
            "Submitted Databricks run '%s' for step '%s', but the run "
            "was not found while retrieving additional metadata.",
            run_id,
            info.pipeline_step_name,
        )
    else:
        extra_metadata: Dict[str, MetadataType] = {}
        if run.job_id is not None:
            extra_metadata[DATABRICKS_STEP_JOB_ID_METADATA_KEY] = str(
                run.job_id
            )
        if run.run_page_url:
            extra_metadata[DATABRICKS_STEP_RUN_URL_METADATA_KEY] = (
                run.run_page_url
            )
        if extra_metadata:
            publish_step_run_metadata(
                info.step_run_id, {self.id: extra_metadata}
            )
            info.step_run.run_metadata.update(extra_metadata)

    logger.info(
        "Submitted step '%s' to Databricks as run '%s'.",
        info.pipeline_step_name,
        run_id,
    )

Functions Modules

databricks_step_operator_entrypoint_config

Entrypoint configuration for Databricks step operator execution.

Classes

DatabricksStepOperatorEntrypointConfiguration(*args: Any, **kwargs: Any)

Bases: StepOperatorEntrypointConfiguration

Entrypoint configuration for Databricks step operator steps.

Source code in src/zenml/step_operators/step_operator_entrypoint_configuration.py

def __init__(self, *args: Any, **kwargs: Any) -> None:
    """Initialize the step operator entrypoint configuration.

    Args:
        *args: The arguments to pass to the superclass.
        **kwargs: The keyword arguments to pass to the superclass.
    """
    super().__init__(*args, **kwargs)
    self._step_run: Optional["StepRunResponse"] = None

Functions

get_entrypoint_arguments(**kwargs: Any) -> List[str] classmethod

Gets the arguments for the Databricks step operator entrypoint.

Parameters:

Name	Type	Description	Default
**kwargs	Any	Keyword arguments, must include the wheel package.	{}

Returns:

Type	Description
List[str]	The entrypoint arguments.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator_entrypoint_config.py

@classmethod
def get_entrypoint_arguments(
    cls,
    **kwargs: Any,
) -> List[str]:
    """Gets the arguments for the Databricks step operator entrypoint.

    Args:
        **kwargs: Keyword arguments, must include the wheel package.

    Returns:
        The entrypoint arguments.
    """
    arguments = super().get_entrypoint_arguments(**kwargs)
    if WHEEL_PACKAGE_OPTION in kwargs:
        arguments.extend(
            [
                f"--{WHEEL_PACKAGE_OPTION}",
                kwargs[WHEEL_PACKAGE_OPTION],
            ]
        )
    return arguments

get_entrypoint_command() -> List[str] classmethod

Returns the command used by Databricks Python wheel tasks.

Returns:

Type	Description
List[str]	Databricks wheel task entrypoint command.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator_entrypoint_config.py

@classmethod
def get_entrypoint_command(cls) -> List[str]:
    """Returns the command used by Databricks Python wheel tasks.

    Returns:
        Databricks wheel task entrypoint command.
    """
    return ["entrypoint.main"]

get_entrypoint_options() -> Dict[str, bool] classmethod

Gets all options required for running with this configuration.

Returns:

Type	Description
Dict[str, bool]	The superclass options plus the wheel package option.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator_entrypoint_config.py

@classmethod
def get_entrypoint_options(cls) -> Dict[str, bool]:
    """Gets all options required for running with this configuration.

    Returns:
        The superclass options plus the wheel package option.
    """
    return super().get_entrypoint_options() | {
        WHEEL_PACKAGE_OPTION: False,
    }

run() -> None

Runs the Databricks step operator step.

Source code in src/zenml/integrations/databricks/step_operators/databricks_step_operator_entrypoint_config.py

def run(self) -> None:
    """Runs the Databricks step operator step."""
    wheel_package = self.entrypoint_args.get(WHEEL_PACKAGE_OPTION)
    if wheel_package:
        configure_databricks_wheel_environment(wheel_package)
    super().run()

Functions

utils

Utilities for Databricks integration.

Modules

databricks_utils

Databricks utilities.

Classes Functions

add_wheel_package_to_sys_path(wheel_package: str) -> str

Add the generated wheel package root to the Python path.

Parameters:

Name	Type	Description	Default
wheel_package	str	The generated wheel package name.	required

Returns:

Type	Description
str	Absolute path to the package root inside the installed wheel.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def add_wheel_package_to_sys_path(wheel_package: str) -> str:
    """Add the generated wheel package root to the Python path.

    Args:
        wheel_package: The generated wheel package name.

    Returns:
        Absolute path to the package root inside the installed wheel.
    """
    project_root = _get_wheel_package_project_root(wheel_package)

    if project_root not in sys.path:
        sys.path.insert(0, project_root)

    return project_root

build_access_control_list(settings: DatabricksBaseSettings) -> Optional[List[iam.AccessControlRequest]]

Build access control entries for Databricks jobs and runs.

Parameters:

Name	Type	Description	Default
settings	DatabricksBaseSettings	Databricks settings.	required

Returns:

Type	Description
Optional[List[AccessControlRequest]]	Access control list or None.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def build_access_control_list(
    settings: DatabricksBaseSettings,
) -> Optional[List[iam.AccessControlRequest]]:
    """Build access control entries for Databricks jobs and runs.

    Args:
        settings: Databricks settings.

    Returns:
        Access control list or None.
    """
    if not settings.access_control_list:
        return None

    return [
        iam.AccessControlRequest(
            group_name=acl.group_name,
            user_name=acl.user_name,
            service_principal_name=acl.service_principal_name,
            permission_level=iam.PermissionLevel(acl.permission_level.value),
        )
        for acl in settings.access_control_list
    ]

build_databricks_cluster_spec(databricks_client: DatabricksClient, settings: DatabricksBaseSettings, env_vars: Dict[str, str]) -> ClusterSpec

Build a Databricks cluster spec from shared settings.

Parameters:

Name	Type	Description	Default
databricks_client	WorkspaceClient	Databricks client.	required
settings	DatabricksBaseSettings	Databricks settings.	required
env_vars	Dict[str, str]	Environment variables for the cluster.	required

Returns:

Type	Description
ClusterSpec	Cluster spec.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def build_databricks_cluster_spec(
    databricks_client: DatabricksClient,
    settings: DatabricksBaseSettings,
    env_vars: Dict[str, str],
) -> ClusterSpec:
    """Build a Databricks cluster spec from shared settings.

    Args:
        databricks_client: Databricks client.
        settings: Databricks settings.
        env_vars: Environment variables for the cluster.

    Returns:
        Cluster spec.
    """
    docker_image = None
    if settings.docker_image_url:
        basic_auth = None
        if settings.docker_image_username:
            basic_auth = DockerBasicAuth(
                username=settings.docker_image_username,
                password=settings.docker_image_password,
            )
        docker_image = DockerImage(
            url=settings.docker_image_url,
            basic_auth=basic_auth,
        )

    init_scripts = None
    if settings.init_scripts:
        init_scripts = [
            _build_init_script_info(script) for script in settings.init_scripts
        ]

    aws_attributes, azure_attributes, gcp_attributes = (
        _get_cloud_availability_attributes(
            databricks_client=databricks_client,
            availability_type=settings.availability_type,
        )
    )

    num_workers = None
    autoscale = None
    if settings.num_workers is not None:
        num_workers = settings.num_workers
    else:
        autoscale = AutoScale(
            min_workers=settings.autoscale[0],
            max_workers=settings.autoscale[1],
        )

    return ClusterSpec(
        spark_version=settings.spark_version
        or DATABRICKS_SPARK_DEFAULT_VERSION,
        node_type_id=settings.node_type_id or DATABRICKS_DEFAULT_NODE_TYPE_ID,
        driver_node_type_id=settings.driver_node_type_id,
        policy_id=_resolve_policy_id(databricks_client, settings.policy_id),
        single_user_name=settings.single_user_name,
        spark_env_vars=env_vars,
        spark_conf=settings.spark_conf or {},
        workload_type=WorkloadType(
            clients=ClientsTypes(jobs=True, notebooks=False)
        ),
        custom_tags=settings.custom_tags,
        docker_image=docker_image,
        init_scripts=init_scripts,
        autotermination_minutes=settings.autotermination_minutes,
        num_workers=num_workers,
        autoscale=autoscale,
        aws_attributes=aws_attributes,
        azure_attributes=azure_attributes,
        gcp_attributes=gcp_attributes,
    )

collect_requirements(docker_settings: DockerSettings, stack: Stack) -> List[str]

Collect clean Python requirements for Databricks wheel execution.

Parameters:

Name	Type	Description	Default
docker_settings	DockerSettings	Docker settings attached to the step or pipeline.	required
stack	Stack	The active stack.	required

Returns:

Type	Description
List[str]	A sorted list of clean requirements.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def collect_requirements(
    docker_settings: "DockerSettings", stack: "Stack"
) -> List[str]:
    """Collect clean Python requirements for Databricks wheel execution.

    Args:
        docker_settings: Docker settings attached to the step or pipeline.
        stack: The active stack.

    Returns:
        A sorted list of clean requirements.
    """
    docker_image_builder = PipelineDockerImageBuilder()
    requirements_files = docker_image_builder.gather_requirements_files(
        docker_settings=docker_settings,
        stack=stack,
        log=False,
    )
    ignored_pip_options = {
        filename: pip_options
        for filename, _, pip_options in requirements_files
        if pip_options
    }
    if ignored_pip_options:
        logger.warning(
            "Databricks PyPI libraries do not support arbitrary pip "
            "options from requirements files. The following pip options "
            "will not be applied: %s",
            ignored_pip_options,
        )

    requirements = [
        line
        for _, content, _ in requirements_files
        for line in content.strip().split("\n")
        if line
    ]
    pip_option_requirements = [
        requirement
        for requirement in requirements
        if _is_pip_option(requirement)
    ]
    if pip_option_requirements:
        logger.warning(
            "Databricks PyPI libraries do not support pip options inside "
            "requirements files. The following requirement lines will not "
            "be applied: %s",
            pip_option_requirements,
        )
        requirements = [
            requirement
            for requirement in requirements
            if not _is_pip_option(requirement)
        ]

    return clean_requirements(sorted(set(requirements)))

configure_databricks_wheel_environment(wheel_package: str) -> str

Configure runtime environment variables for a Databricks wheel.

Parameters:

Name	Type	Description	Default
wheel_package	str	The generated wheel package name.	required

Returns:

Type	Description
str	Absolute path to the package root inside the installed wheel.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def configure_databricks_wheel_environment(wheel_package: str) -> str:
    """Configure runtime environment variables for a Databricks wheel.

    Args:
        wheel_package: The generated wheel package name.

    Returns:
        Absolute path to the package root inside the installed wheel.
    """
    project_root = add_wheel_package_to_sys_path(wheel_package)
    os.environ[ENV_ZENML_DATABRICKS_WHEEL_PACKAGE] = wheel_package
    os.environ[ENV_ZENML_CUSTOM_SOURCE_ROOT] = project_root
    return project_root

convert_step_to_submit_task(task_name: str, command: str, arguments: List[str], new_cluster: ClusterSpec, libraries: Optional[List[str]] = None, zenml_project_wheel: Optional[str] = None, timeout_seconds: Optional[int] = None) -> SubmitTask

Convert a ZenML step to a Databricks submit task.

Parameters:

Name	Type	Description	Default
task_name	str	Name of the task.	required
command	str	Command to run.	required
arguments	List[str]	Arguments to pass to the command.	required
new_cluster	ClusterSpec	Cluster spec for the submit task.	required
libraries	Optional[List[str]]	List of libraries to install.	None
zenml_project_wheel	Optional[str]	Path to the ZenML project wheel.	None
timeout_seconds	Optional[int]	Timeout in seconds for the task.	None

Returns:

Type	Description
SubmitTask	Databricks submit task.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def convert_step_to_submit_task(
    task_name: str,
    command: str,
    arguments: List[str],
    new_cluster: ClusterSpec,
    libraries: Optional[List[str]] = None,
    zenml_project_wheel: Optional[str] = None,
    timeout_seconds: Optional[int] = None,
) -> SubmitTask:
    """Convert a ZenML step to a Databricks submit task.

    Args:
        task_name: Name of the task.
        command: Command to run.
        arguments: Arguments to pass to the command.
        new_cluster: Cluster spec for the submit task.
        libraries: List of libraries to install.
        zenml_project_wheel: Path to the ZenML project wheel.
        timeout_seconds: Timeout in seconds for the task.

    Returns:
        Databricks submit task.
    """
    return SubmitTask(
        task_key=task_name,
        libraries=_get_databricks_libraries(
            libraries=libraries,
            zenml_project_wheel=zenml_project_wheel,
        ),
        new_cluster=new_cluster,
        python_wheel_task=PythonWheelTask(
            package_name="zenml",
            entry_point=command,
            parameters=arguments,
        ),
        timeout_seconds=timeout_seconds,
    )

convert_step_to_task(task_name: str, command: str, arguments: List[str], libraries: Optional[List[str]] = None, depends_on: Optional[List[str]] = None, zenml_project_wheel: Optional[str] = None, job_cluster_key: Optional[str] = None, timeout_seconds: Optional[int] = None, max_retries: Optional[int] = None, min_retry_interval_millis: Optional[int] = None, retry_on_timeout: Optional[bool] = None) -> DatabricksTask

Convert a ZenML step to a Databricks task.

Parameters:

Name	Type	Description	Default
task_name	str	Name of the task.	required
command	str	Command to run.	required
arguments	List[str]	Arguments to pass to the command.	required
libraries	Optional[List[str]]	List of libraries to install.	None
depends_on	Optional[List[str]]	List of tasks to depend on.	None
zenml_project_wheel	Optional[str]	Path to the ZenML project wheel.	None
job_cluster_key	Optional[str]	ID of the Databricks job_cluster_key.	None
timeout_seconds	Optional[int]	Timeout in seconds for the task.	None
max_retries	Optional[int]	Maximum number of retries for a failed task.	None
min_retry_interval_millis	Optional[int]	Minimum interval between retries in milliseconds.	None
retry_on_timeout	Optional[bool]	Whether to retry on timeout.	None

Returns:

Type	Description
Task	Databricks task.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def convert_step_to_task(
    task_name: str,
    command: str,
    arguments: List[str],
    libraries: Optional[List[str]] = None,
    depends_on: Optional[List[str]] = None,
    zenml_project_wheel: Optional[str] = None,
    job_cluster_key: Optional[str] = None,
    timeout_seconds: Optional[int] = None,
    max_retries: Optional[int] = None,
    min_retry_interval_millis: Optional[int] = None,
    retry_on_timeout: Optional[bool] = None,
) -> DatabricksTask:
    """Convert a ZenML step to a Databricks task.

    Args:
        task_name: Name of the task.
        command: Command to run.
        arguments: Arguments to pass to the command.
        libraries: List of libraries to install.
        depends_on: List of tasks to depend on.
        zenml_project_wheel: Path to the ZenML project wheel.
        job_cluster_key: ID of the Databricks job_cluster_key.
        timeout_seconds: Timeout in seconds for the task.
        max_retries: Maximum number of retries for a failed task.
        min_retry_interval_millis: Minimum interval between retries
            in milliseconds.
        retry_on_timeout: Whether to retry on timeout.

    Returns:
        Databricks task.
    """
    return DatabricksTask(
        task_key=task_name,
        job_cluster_key=job_cluster_key,
        libraries=_get_databricks_libraries(
            libraries=libraries,
            zenml_project_wheel=zenml_project_wheel,
        ),
        python_wheel_task=PythonWheelTask(
            package_name="zenml",
            entry_point=command,
            parameters=arguments,
        ),
        depends_on=[TaskDependency(task) for task in depends_on]
        if depends_on
        else None,
        timeout_seconds=timeout_seconds,
        max_retries=max_retries,
        min_retry_interval_millis=min_retry_interval_millis,
        retry_on_timeout=retry_on_timeout,
    )

delete_workspace_directory(databricks_client: DatabricksClient, databricks_directory: str, context: str) -> None

Delete a Databricks workspace directory best-effort.

Parameters:

Name	Type	Description	Default
databricks_client	WorkspaceClient	Databricks client.	required
databricks_directory	str	Workspace directory to delete.	required
context	str	Context for warning logs if cleanup fails.	required

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def delete_workspace_directory(
    databricks_client: DatabricksClient,
    databricks_directory: str,
    context: str,
) -> None:
    """Delete a Databricks workspace directory best-effort.

    Args:
        databricks_client: Databricks client.
        databricks_directory: Workspace directory to delete.
        context: Context for warning logs if cleanup fails.
    """
    try:
        databricks_client.workspace.delete(
            path=databricks_directory,
            recursive=True,
        )
    except NotFound:
        logger.debug(
            "Databricks workspace wheel directory `%s` was already "
            "deleted after %s.",
            databricks_directory,
            context,
        )
    except Exception as e:
        logger.warning(
            "Failed to delete Databricks workspace wheel directory `%s` "
            "after %s: %s",
            databricks_directory,
            context,
            e,
        )

get_databricks_wheel_source() -> Optional[tuple[str, str]]

Get the source root and package name of an installed Databricks wheel.

When code is already running from a Databricks wheel, building a new wheel from the active custom source root can point at the Databricks working directory instead of the packaged project source. In that case we rebuild from the installed wheel package contents instead.

Returns:

Type	Description
Optional[tuple[str, str]]	Tuple of (source_root, package_name) if running from a Databricks wheel,
Optional[tuple[str, str]]	otherwise None.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def get_databricks_wheel_source() -> Optional[tuple[str, str]]:
    """Get the source root and package name of an installed Databricks wheel.

    When code is already running from a Databricks wheel, building a new wheel
    from the active custom source root can point at the Databricks working
    directory instead of the packaged project source. In that case we rebuild
    from the installed wheel package contents instead.

    Returns:
        Tuple of (source_root, package_name) if running from a Databricks wheel,
        otherwise None.
    """
    wheel_package = os.environ.get(ENV_ZENML_DATABRICKS_WHEEL_PACKAGE)
    if not wheel_package:
        return None

    return _get_wheel_package_project_root(wheel_package), wheel_package

map_databricks_run_to_execution_status(run: Run) -> ExecutionStatus

Map a Databricks run state to a ZenML execution status.

Parameters:

Name	Type	Description	Default
run	Run	Databricks run.	required

Returns:

Type	Description
ExecutionStatus	ZenML execution status.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def map_databricks_run_to_execution_status(run: Run) -> ExecutionStatus:
    """Map a Databricks run state to a ZenML execution status.

    Args:
        run: Databricks run.

    Returns:
        ZenML execution status.
    """
    state = run.state
    if state is None or state.life_cycle_state is None:
        return ExecutionStatus.FAILED

    life_cycle_state = state.life_cycle_state
    if life_cycle_state in {
        RunLifeCycleState.BLOCKED,
        RunLifeCycleState.PENDING,
        RunLifeCycleState.QUEUED,
        RunLifeCycleState.RUNNING,
        RunLifeCycleState.TERMINATING,
    }:
        return ExecutionStatus.RUNNING

    if life_cycle_state == RunLifeCycleState.WAITING_FOR_RETRY:
        return ExecutionStatus.RUNNING

    if life_cycle_state in {
        RunLifeCycleState.INTERNAL_ERROR,
        RunLifeCycleState.SKIPPED,
    }:
        return ExecutionStatus.FAILED

    if life_cycle_state != RunLifeCycleState.TERMINATED:
        return ExecutionStatus.FAILED

    result_state = state.result_state
    if result_state in {
        RunResultState.SUCCESS,
        RunResultState.SUCCESS_WITH_FAILURES,
    }:
        return ExecutionStatus.COMPLETED

    if result_state in {
        RunResultState.CANCELED,
        RunResultState.UPSTREAM_CANCELED,
    }:
        return ExecutionStatus.STOPPED

    return ExecutionStatus.FAILED

sanitize_labels(labels: Dict[str, str]) -> None

Update the label values to be valid Databricks tag values.

Databricks tag values must contain only alphanumeric characters, underscores, hyphens, and periods, and must be at most 63 characters long with leading/trailing separators trimmed.

Parameters:

Name	Type	Description	Default
labels	Dict[str, str]	the labels to sanitize.	required

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def sanitize_labels(labels: Dict[str, str]) -> None:
    """Update the label values to be valid Databricks tag values.

    Databricks tag values must contain only alphanumeric characters,
    underscores, hyphens, and periods, and must be at most 63 characters
    long with leading/trailing separators trimmed.

    Args:
        labels: the labels to sanitize.
    """
    for key, value in labels.items():
        labels[key] = re.sub(r"[^0-9a-zA-Z-_\.]+", "_", value)[:63].strip(
            "-_."
        )

upload_wheel_to_workspace(databricks_client: DatabricksClient, wheel_path: str, databricks_directory: str) -> str

Upload a wheel file to a Databricks workspace directory.

Parameters:

Name	Type	Description	Default
databricks_client	WorkspaceClient	Databricks client.	required
wheel_path	str	Local wheel path.	required
databricks_directory	str	Remote Databricks directory path.	required

Raises:

Type	Description
RuntimeError	If the wheel upload fails.

Returns:

Type	Description
str	The Databricks workspace wheel path.

Source code in src/zenml/integrations/databricks/utils/databricks_utils.py

def upload_wheel_to_workspace(
    databricks_client: DatabricksClient,
    wheel_path: str,
    databricks_directory: str,
) -> str:
    """Upload a wheel file to a Databricks workspace directory.

    Args:
        databricks_client: Databricks client.
        wheel_path: Local wheel path.
        databricks_directory: Remote Databricks directory path.

    Raises:
        RuntimeError: If the wheel upload fails.

    Returns:
        The Databricks workspace wheel path.
    """
    wheel_filename = os.path.basename(wheel_path)
    databricks_wheel_path = f"{databricks_directory}/{wheel_filename}"

    try:
        databricks_client.workspace.mkdirs(path=databricks_directory)
        with open(wheel_path, "rb") as f:
            databricks_client.workspace.upload(
                path=databricks_wheel_path,
                content=f.read(),
                format=ImportFormat.AUTO,
                overwrite=True,
            )
    except Exception as e:
        raise RuntimeError(
            f"Failed to upload wheel file to Databricks workspace at "
            f"{databricks_wheel_path}. Ensure your Databricks workspace has "
            f"the necessary permissions and the path is accessible. "
            f"Original error: {e}"
        ) from e

    logger.info(
        "Successfully uploaded wheel to Databricks workspace: %s",
        databricks_wheel_path,
    )
    return databricks_wheel_path