zenml-nightly 0.70.0.dev20241130__tar.gz → 0.70.0.dev20241203__tar.gz

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: zenml-nightly
-Version: 0.70.0.dev20241130
+Version: 0.70.0.dev20241203
 Summary: ZenML: Write production-ready ML code.
 Home-page: https://zenml.io
 License: Apache-2.0

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/pyproject.toml

@@ -1,6 +1,6 @@
 [tool.poetry]
 name = "zenml-nightly"
-version = "0.70.0.dev20241130"
+version = "0.70.0.dev20241203"
 packages = [{ include = "zenml", from = "src" }]
 description = "ZenML: Write production-ready ML code."
 authors = ["ZenML GmbH <info@zenml.io>"]

zenml_nightly-0.70.0.dev20241203/src/zenml/VERSION

@@ -0,0 +1 @@
+0.70.0.dev20241203

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/cli/__init__.py

@@ -1987,7 +1987,7 @@ Secrets management
 ------------------
 
 ZenML offers a way to [securely store secrets associated with your other
-stack components and infrastructure](https://docs.zenml.io/getting-started/deploying-zenml/manage-the-deployed-services/secret-management).
+stack components and infrastructure](https://docs.zenml.io/getting-started/deploying-zenml/secret-management).
 A ZenML Secret is a collection or grouping of key-value pairs stored by the
 ZenML secrets store. ZenML Secrets are identified by a unique name which
 allows you to fetch or reference them in your pipelines and stacks.
@@ -2083,7 +2083,7 @@ challenge in configuring uninterrupted, secure access to infrastructure
 resources. In ZenML, Service Connectors streamline this process by abstracting
 away the complexity of authentication and help you connect your stack to your
 resources. You can find the full docs on the ZenML service connectors
-[here](https://docs.zenml.io/how-to/auth-management).
+[here](https://docs.zenml.io/how-to/infrastructure-deployment/auth-management).
 
 The ZenML CLI features a variety of commands to help you manage your service
 connectors. First of all, to explore all the types of service connectors
@@ -2113,7 +2113,7 @@ zenml service-connector register SERVICE_CONNECTOR_NAME \
 ```
 
 For more details on how to create a service connector, please refer to our
-[docs](https://docs.zenml.io/how-to/auth-management).
+[docs](https://docs.zenml.io/how-to/infrastructure-deployment/auth-management).
 
 To check if your service connector is registered properly, you can `verify` it.
 By doing this, you can both check if it is configured correctly and also, you
@@ -2367,7 +2367,7 @@ defining the pipeline is not in your current directory, the module path consists
 of the full path to the file, separated by dots, e.g.
 `some_directory.some_file.my_pipeline`.
 
-To [build Docker images for your pipeline](https://docs.zenml.io/how-to/customize-docker-builds)
+To [build Docker images for your pipeline](https://docs.zenml.io/how-to/infrastructure-deployment/customize-docker-builds)
 without actually running the pipeline, use:
 
 ```bash

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/cli/base.py

@@ -264,7 +264,7 @@ def init(
         f"will only take effect when you're running ZenML from the initialized "
         f"repository root, or from a subdirectory. For more information on "
         f"repositories and configurations, please visit "
-        f"https://docs.zenml.io/user-guide/starter-guide/understand-stacks."
+        f"https://docs.zenml.io/user-guide/production-guide/understand-stacks."
     )
 
 

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/cli/pipeline.py

@@ -40,6 +40,35 @@ from zenml.utils.yaml_utils import write_yaml
 logger = get_logger(__name__)
 
 
+def _import_pipeline(source: str) -> Pipeline:
+    """Import a pipeline.
+
+    Args:
+        source: The pipeline source.
+
+    Returns:
+        The pipeline.
+    """
+    try:
+        pipeline_instance = source_utils.load(source)
+    except ModuleNotFoundError as e:
+        source_root = source_utils.get_source_root()
+        cli_utils.error(
+            f"Unable to import module `{e.name}`. Make sure the source path is "
+            f"relative to your source root `{source_root}`."
+        )
+    except AttributeError as e:
+        cli_utils.error("Unable to load attribute from module: " + str(e))
+
+    if not isinstance(pipeline_instance, Pipeline):
+        cli_utils.error(
+            f"The given source path `{source}` does not resolve to a pipeline "
+            "object."
+        )
+
+    return pipeline_instance
+
+
 @cli.group(cls=TagGroup, tag=CliCategories.MANAGEMENT_TOOLS)
 def pipeline() -> None:
     """Interact with pipelines, runs and schedules."""
@@ -85,22 +114,7 @@ def register_pipeline(
             "source code root."
         )
 
-    try:
-        pipeline_instance = source_utils.load(source)
-    except ModuleNotFoundError as e:
-        source_root = source_utils.get_source_root()
-        cli_utils.error(
-            f"Unable to import module `{e.name}`. Make sure the source path is "
-            f"relative to your source root `{source_root}`."
-        )
-    except AttributeError as e:
-        cli_utils.error("Unable to load attribute from module: " + str(e))
-
-    if not isinstance(pipeline_instance, Pipeline):
-        cli_utils.error(
-            f"The given source path `{source}` does not resolve to a pipeline "
-            "object."
-        )
+    pipeline_instance = _import_pipeline(source=source)
 
     parameters: Dict[str, Any] = {}
     if parameters_path:
@@ -176,24 +190,9 @@ def build_pipeline(
             "your source code root."
         )
 
-    try:
-        pipeline_instance = source_utils.load(source)
-    except ModuleNotFoundError as e:
-        source_root = source_utils.get_source_root()
-        cli_utils.error(
-            f"Unable to import module `{e.name}`. Make sure the source path is "
-            f"relative to your source root `{source_root}`."
-        )
-    except AttributeError as e:
-        cli_utils.error("Unable to load attribute from module: " + str(e))
-
-    if not isinstance(pipeline_instance, Pipeline):
-        cli_utils.error(
-            f"The given source path `{source}` does not resolve to a pipeline "
-            "object."
-        )
-
     with cli_utils.temporary_active_stack(stack_name_or_id=stack_name_or_id):
+        pipeline_instance = _import_pipeline(source=source)
+
         pipeline_instance = pipeline_instance.with_options(
             config_path=config_path
         )
@@ -277,36 +276,21 @@ def run_pipeline(
             "your source code root."
         )
 
-    try:
-        pipeline_instance = source_utils.load(source)
-    except ModuleNotFoundError as e:
-        source_root = source_utils.get_source_root()
-        cli_utils.error(
-            f"Unable to import module `{e.name}`. Make sure the source path is "
-            f"relative to your source root `{source_root}`."
-        )
-    except AttributeError as e:
-        cli_utils.error("Unable to load attribute from module: " + str(e))
-
-    if not isinstance(pipeline_instance, Pipeline):
-        cli_utils.error(
-            f"The given source path `{source}` does not resolve to a pipeline "
-            "object."
-        )
-
-    build: Union[str, PipelineBuildBase, None] = None
-    if build_path_or_id:
-        if uuid_utils.is_valid_uuid(build_path_or_id):
-            build = build_path_or_id
-        elif os.path.exists(build_path_or_id):
-            build = PipelineBuildBase.from_yaml(build_path_or_id)
-        else:
-            cli_utils.error(
-                f"The specified build {build_path_or_id} is not a valid UUID "
-                "or file path."
-            )
-
     with cli_utils.temporary_active_stack(stack_name_or_id=stack_name_or_id):
+        pipeline_instance = _import_pipeline(source=source)
+
+        build: Union[str, PipelineBuildBase, None] = None
+        if build_path_or_id:
+            if uuid_utils.is_valid_uuid(build_path_or_id):
+                build = build_path_or_id
+            elif os.path.exists(build_path_or_id):
+                build = PipelineBuildBase.from_yaml(build_path_or_id)
+            else:
+                cli_utils.error(
+                    f"The specified build {build_path_or_id} is not a valid UUID "
+                    "or file path."
+                )
+
         pipeline_instance = pipeline_instance.with_options(
             config_path=config_path,
             build=build,
@@ -369,24 +353,9 @@ def create_run_template(
             "init` at your source code root."
         )
 
-    try:
-        pipeline_instance = source_utils.load(source)
-    except ModuleNotFoundError as e:
-        source_root = source_utils.get_source_root()
-        cli_utils.error(
-            f"Unable to import module `{e.name}`. Make sure the source path is "
-            f"relative to your source root `{source_root}`."
-        )
-    except AttributeError as e:
-        cli_utils.error("Unable to load attribute from module: " + str(e))
-
-    if not isinstance(pipeline_instance, Pipeline):
-        cli_utils.error(
-            f"The given source path `{source}` does not resolve to a pipeline "
-            "object."
-        )
-
     with cli_utils.temporary_active_stack(stack_name_or_id=stack_name_or_id):
+        pipeline_instance = _import_pipeline(source=source)
+
         pipeline_instance = pipeline_instance.with_options(
             config_path=config_path
         )

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/config/secret_reference_mixin.py

@@ -66,7 +66,7 @@ class SecretReferenceMixin(BaseModel):
                         "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/getting-started/deploying-zenml/manage-the-deployed-services/secret-management"
+                        "here: https://docs.zenml.io/getting-started/deploying-zenml/secret-management"
                     )
                 continue
 

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/integrations/gcp/orchestrators/vertex_orchestrator.py

@@ -793,7 +793,7 @@ class VertexOrchestrator(ContainerizedOrchestrator, GoogleCredentialsMixin):
                     "set or set to 0. The accelerator type will be ignored. "
                     "To fix this warning, either remove the specified "
                     "accelerator type or set the `gpu_count` using the "
-                    "ResourceSettings (https://docs.zenml.io/how-to/training-with-gpus#specify-resource-requirements-for-steps)."
+                    "ResourceSettings (https://docs.zenml.io/how-to/advanced-topics/training-with-gpus)."
                 )
 
         return dynamic_component

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/integrations/neptune/experiment_trackers/neptune_experiment_tracker.py

@@ -77,10 +77,12 @@ class NeptuneExperimentTracker(BaseExperimentTracker):
             NeptuneExperimentTrackerSettings, self.get_settings(info)
         )
 
-        self.run_state.token = self.config.api_token
-        self.run_state.project = self.config.project
-        self.run_state.run_name = info.run_name
-        self.run_state.tags = list(settings.tags)
+        self.run_state.initialize(
+            project=self.config.project,
+            token=self.config.api_token,
+            run_name=info.run_name,
+            tags=list(settings.tags),
+        )
 
     def get_step_run_metadata(
         self, info: "StepRunInfo"
@@ -107,4 +109,4 @@ class NeptuneExperimentTracker(BaseExperimentTracker):
         """
         self.run_state.active_run.sync()
         self.run_state.active_run.stop()
-        self.run_state.reset_active_run()
+        self.run_state.reset()

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/integrations/neptune/experiment_trackers/run_state.py

@@ -20,7 +20,6 @@ import neptune
 
 import zenml
 from zenml.client import Client
-from zenml.integrations.constants import NEPTUNE
 from zenml.utils.singleton import SingletonMetaClass
 
 if TYPE_CHECKING:
@@ -29,20 +28,38 @@ if TYPE_CHECKING:
 _INTEGRATION_VERSION_KEY = "source_code/integrations/zenml"
 
 
-class InvalidExperimentTrackerSelected(Exception):
-    """Raised if a Neptune run is fetched while using a different experiment tracker."""
-
-
 class RunProvider(metaclass=SingletonMetaClass):
     """Singleton object used to store and persist a Neptune run state across the pipeline."""
 
     def __init__(self) -> None:
         """Initialize RunProvider. Called with no arguments."""
         self._active_run: Optional["Run"] = None
-        self._project: Optional[str]
-        self._run_name: Optional[str]
-        self._token: Optional[str]
-        self._tags: Optional[List[str]]
+        self._project: Optional[str] = None
+        self._run_name: Optional[str] = None
+        self._token: Optional[str] = None
+        self._tags: Optional[List[str]] = None
+        self._initialized = False
+
+    def initialize(
+        self,
+        project: Optional[str] = None,
+        token: Optional[str] = None,
+        run_name: Optional[str] = None,
+        tags: Optional[List[str]] = None,
+    ) -> None:
+        """Initialize the run state.
+
+        Args:
+            project: The neptune project.
+            token: The neptune token.
+            run_name: The neptune run name.
+            tags: Tags for the neptune run.
+        """
+        self._project = project
+        self._token = token
+        self._run_name = run_name
+        self._tags = tags
+        self._initialized = True
 
     @property
     def project(self) -> Optional[Any]:
@@ -53,15 +70,6 @@ class RunProvider(metaclass=SingletonMetaClass):
         """
         return self._project
 
-    @project.setter
-    def project(self, project: str) -> None:
-        """Setter for project name.
-
-        Args:
-            project: Neptune project name
-        """
-        self._project = project
-
     @property
     def token(self) -> Optional[Any]:
         """Getter for API token.
@@ -71,15 +79,6 @@ class RunProvider(metaclass=SingletonMetaClass):
         """
         return self._token
 
-    @token.setter
-    def token(self, token: str) -> None:
-        """Setter for API token.
-
-        Args:
-            token: Neptune API token
-        """
-        self._token = token
-
     @property
     def run_name(self) -> Optional[Any]:
         """Getter for run name.
@@ -89,15 +88,6 @@ class RunProvider(metaclass=SingletonMetaClass):
         """
         return self._run_name
 
-    @run_name.setter
-    def run_name(self, run_name: str) -> None:
-        """Setter for run name.
-
-        Args:
-            run_name: name of the pipeline run
-        """
-        self._run_name = run_name
-
     @property
     def tags(self) -> Optional[Any]:
         """Getter for run tags.
@@ -107,14 +97,14 @@ class RunProvider(metaclass=SingletonMetaClass):
         """
         return self._tags
 
-    @tags.setter
-    def tags(self, tags: List[str]) -> None:
-        """Setter for run tags.
+    @property
+    def initialized(self) -> bool:
+        """If the run state is initialized.
 
-        Args:
-            tags: list of tags associated with a Neptune run
+        Returns:
+            If the run state is initialized.
         """
-        self._tags = tags
+        return self._initialized
 
     @property
     def active_run(self) -> "Run":
@@ -137,9 +127,14 @@ class RunProvider(metaclass=SingletonMetaClass):
             self._active_run = run
         return self._active_run
 
-    def reset_active_run(self) -> None:
-        """Resets the active run state to None."""
+    def reset(self) -> None:
+        """Reset the run state."""
         self._active_run = None
+        self._project = None
+        self._run_name = None
+        self._token = None
+        self._tags = None
+        self._initialized = False
 
 
 def get_neptune_run() -> "Run":
@@ -149,14 +144,35 @@ def get_neptune_run() -> "Run":
         Neptune run object
 
     Raises:
-        InvalidExperimentTrackerSelected: when called while using an experiment tracker other than Neptune
+        RuntimeError: When unable to fetch the active neptune run.
     """
-    client = Client()
-    experiment_tracker = client.active_stack.experiment_tracker
-    if experiment_tracker.flavor == NEPTUNE:  # type: ignore
-        return experiment_tracker.run_state.active_run  # type: ignore
-    raise InvalidExperimentTrackerSelected(
-        "Fetching a Neptune run works only with the 'neptune' flavor of "
-        "the experiment tracker. The flavor currently selected is %s"
-        % experiment_tracker.flavor  # type: ignore
+    from zenml.integrations.neptune.experiment_trackers import (
+        NeptuneExperimentTracker,
     )
+
+    experiment_tracker = Client().active_stack.experiment_tracker
+
+    if not experiment_tracker:
+        raise RuntimeError(
+            "Unable to get neptune run: Missing experiment tracker in the "
+            "active stack."
+        )
+
+    if not isinstance(experiment_tracker, NeptuneExperimentTracker):
+        raise RuntimeError(
+            "Unable to get neptune run: Experiment tracker in the active "
+            f"stack ({experiment_tracker.flavor}) is not a neptune experiment "
+            "tracker."
+        )
+
+    run_state = experiment_tracker.run_state
+    if not run_state.initialized:
+        raise RuntimeError(
+            "Unable to get neptune run: The experiment tracker has not been "
+            "initialized. To solve this, make sure you use the experiment "
+            "tracker in your step. See "
+            "https://docs.zenml.io/stack-components/experiment-trackers/neptune#how-do-you-use-it "
+            "for more information."
+        )
+
+    return experiment_tracker.run_state.active_run

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/integrations/registry.py

@@ -111,7 +111,7 @@ class IntegrationRegistry(object):
                 )
             else:
                 raise KeyError(
-                    f"Version {integration_name} does not exist. "
+                    f"Integration {integration_name} does not exist. "
                     f"Currently the following integrations are implemented. "
                     f"{self.list_integration_names}"
                 )
@@ -148,7 +148,7 @@ class IntegrationRegistry(object):
                 ].get_uninstall_requirements(target_os=target_os)
             else:
                 raise KeyError(
-                    f"Version {integration_name} does not exist. "
+                    f"Integration {integration_name} does not exist. "
                     f"Currently the following integrations are implemented. "
                     f"{self.list_integration_names}"
                 )

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/orchestrators/base_orchestrator.py

@@ -61,7 +61,7 @@ class BaseOrchestratorConfig(StackComponentConfig):
                     "The 'custom_docker_base_image_name' field has been "
                     "deprecated. To use a custom base container image with your "
                     "orchestrators, please use the DockerSettings in your "
-                    "pipeline (see https://docs.zenml.io/how-to/customize-docker-builds)."
+                    "pipeline (see https://docs.zenml.io/how-to/infrastructure-deployment/customize-docker-builds)."
                 )
 
         return data

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/service_connectors/service_connector_utils.py

@@ -61,13 +61,13 @@ def _raise_specific_cloud_exception_if_needed(
     container_registries: List[ResourcesInfo],
 ) -> None:
     AWS_DOCS = (
-        "https://docs.zenml.io/how-to/auth-management/aws-service-connector"
+        "https://docs.zenml.io/how-to/infrastructure-deployment/auth-management/aws-service-connector"
     )
     GCP_DOCS = (
-        "https://docs.zenml.io/how-to/auth-management/gcp-service-connector"
+        "https://docs.zenml.io/how-to/infrastructure-deployment/auth-management/gcp-service-connector"
     )
     AZURE_DOCS = (
-        "https://docs.zenml.io/how-to/auth-management/azure-service-connector"
+        "https://docs.zenml.io/how-to/infrastructure-deployment/auth-management/azure-service-connector"
     )
 
     if not artifact_stores:

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/stack/stack_component.py

@@ -102,7 +102,7 @@ class StackComponentConfig(BaseModel, ABC):
                         "in sensitive information as secrets. Check out the "
                         "documentation on how to configure your stack "
                         "components with secrets here: "
-                        "https://docs.zenml.io/getting-started/deploying-zenml/manage-the-deployed-services/secret-management"
+                        "https://docs.zenml.io/getting-started/deploying-zenml/secret-management"
                     )
                 continue
 

{zenml_nightly-0.70.0.dev20241130 → zenml_nightly-0.70.0.dev20241203}/src/zenml/zen_stores/rest_zen_store.py

@@ -4062,7 +4062,7 @@ class RestZenStore(BaseZenStore):
                 "you should use a service account API key to authenticate to "
                 "the server instead of temporary CLI login credentials. For "
                 "more information, see "
-                "https://docs.zenml.io/how-to/connecting-to-zenml/connect-with-a-service-account"
+                "https://docs.zenml.io/how-to/project-setup-and-management/connecting-to-zenml/connect-with-a-service-account"
             )
 
             if api_key is not None:

zenml_nightly-0.70.0.dev20241130/src/zenml/VERSION

@@ -1 +0,0 @@
-0.70.0.dev20241130