mlrun 1.8.0rc15__py3-none-any.whl → 1.8.0rc17__py3-none-any.whl

mlrun/model_monitoring/writer.py

@@ -31,7 +31,6 @@ from mlrun.common.schemas.model_monitoring.constants import (
     WriterEvent,
     WriterEventKind,
 )
-from mlrun.common.schemas.notification import NotificationKind, NotificationSeverity
 from mlrun.model_monitoring.db._stats import (
     ModelMonitoringCurrentStatsFile,
     ModelMonitoringDriftMeasuresFile,
@@ -39,7 +38,6 @@ from mlrun.model_monitoring.db._stats import (
 from mlrun.model_monitoring.helpers import get_result_instance_fqn
 from mlrun.serving.utils import StepToDict
 from mlrun.utils import logger
-from mlrun.utils.notifications.notification_pusher import CustomNotificationPusher
 
 _RawEvent = dict[str, Any]
 _AppResultEvent = NewType("_AppResultEvent", _RawEvent)
@@ -57,50 +55,6 @@ class _WriterEventTypeError(_WriterEventError, TypeError):
     pass
 
 
-class _Notifier:
-    def __init__(
-        self,
-        event: _AppResultEvent,
-        notification_pusher: CustomNotificationPusher,
-        severity: NotificationSeverity = NotificationSeverity.WARNING,
-    ) -> None:
-        """
-        Event notifier - send push notification when appropriate to the notifiers in
-        `notification pusher`.
-        Note that if you use a Slack App webhook, you need to define it as an MLRun secret
-        `SLACK_WEBHOOK`.
-        """
-        self._event = event
-        self._custom_notifier = notification_pusher
-        self._severity = severity
-
-    def _should_send_event(self) -> bool:
-        return self._event[ResultData.RESULT_STATUS] >= ResultStatusApp.detected.value
-
-    def _generate_message(self) -> str:
-        return f"""\
-The monitoring app `{self._event[WriterEvent.APPLICATION_NAME]}` \
-of kind `{self._event[ResultData.RESULT_KIND]}` \
-detected a problem in model endpoint ID `{self._event[WriterEvent.ENDPOINT_ID]}` \
-at time `{self._event[WriterEvent.START_INFER_TIME]}`.
-
-Result data:
-Name: `{self._event[ResultData.RESULT_NAME]}`
-Value: `{self._event[ResultData.RESULT_VALUE]}`
-Status: `{self._event[ResultData.RESULT_STATUS]}`
-Extra data: `{self._event[ResultData.RESULT_EXTRA_DATA]}`\
-"""
-
-    def notify(self) -> None:
-        """Send notification if appropriate"""
-        if not self._should_send_event():
-            logger.debug("Not sending a notification")
-            return
-        message = self._generate_message()
-        self._custom_notifier.push(message=message, severity=self._severity)
-        logger.debug("A notification should have been sent")
-
-
 class ModelMonitoringWriter(StepToDict):
     """
     Write monitoring application results to the target databases
@@ -116,10 +70,6 @@ class ModelMonitoringWriter(StepToDict):
         self.project = project
         self.name = project  # required for the deployment process
 
-        self._custom_notifier = CustomNotificationPusher(
-            notification_types=[NotificationKind.slack]
-        )
-
         self._tsdb_connector = mlrun.model_monitoring.get_tsdb_connector(
             project=self.project, secret_provider=secret_provider
         )
@@ -250,9 +200,6 @@ class ModelMonitoringWriter(StepToDict):
 
         logger.info("Completed event DB writes")
 
-        if kind == WriterEventKind.RESULT:
-            _Notifier(event=event, notification_pusher=self._custom_notifier).notify()
-
         if (
             mlrun.mlconf.alerts.mode == mlrun.common.schemas.alert.AlertsModes.enabled
             and kind == WriterEventKind.RESULT

mlrun/projects/pipelines.py

@@ -647,7 +647,7 @@ class _KFPRunner(_PipelineRunner):
                     exc_info=err_to_str(exc),
                 )
 
-        # TODO: we should check how can we get the run uid when we don't the the context (for example on
+        # TODO: we should check how can we get the run uid when we don't have the context (for example on
         #  mlrun.load_project() and later call directly to project.run)
         if context:
             project.notifiers.push_pipeline_start_message(

mlrun/projects/project.py

@@ -44,6 +44,7 @@ import mlrun.common.runtimes.constants
 import mlrun.common.schemas.alert
 import mlrun.common.schemas.artifact
 import mlrun.common.schemas.model_monitoring.constants as mm_constants
+import mlrun.datastore.datastore_profile
 import mlrun.db
 import mlrun.errors
 import mlrun.k8s_utils
@@ -1961,6 +1962,11 @@ class MlrunProject(ModelObj):
             ... )
 
         """
+        if not document_loader_spec.download_object and upload:
+            raise ValueError(
+                "This document loader expects direct links/URLs and does not support file uploads. "
+                "Either set download_object=True or set upload=False"
+            )
         doc_artifact = DocumentArtifact(
             key=key,
             original_source=local_path or target_path,
@@ -3579,8 +3585,6 @@ class MlrunProject(ModelObj):
                                           * None - will be set from the system configuration.
                                           * v3io - for v3io endpoint store, pass `v3io` and the system will generate the
                                             exact path.
-                                          * MySQL/SQLite - for SQL endpoint store, provide the full connection string,
-                                            for example: mysql+pymysql://<username>:<password>@<host>:<port>/<db_name>
         :param stream_path:               Path to the model monitoring stream. By default, None. Options:
 
                                           * None - will be set from the system configuration.
@@ -3603,12 +3607,30 @@ class MlrunProject(ModelObj):
                                           & tracked model server.
         """
         db = mlrun.db.get_run_db(secrets=self._secrets)
+        if tsdb_connection == "v3io":
+            tsdb_profile = mlrun.datastore.datastore_profile.DatastoreProfileV3io(
+                name="mm-infra-tsdb"
+            )
+            self.register_datastore_profile(tsdb_profile)
+            tsdb_profile_name = tsdb_profile.name
+        else:
+            tsdb_profile_name = None
+        if stream_path == "v3io":
+            stream_profile = mlrun.datastore.datastore_profile.DatastoreProfileV3io(
+                name="mm-infra-stream"
+            )
+            self.register_datastore_profile(stream_profile)
+            stream_profile_name = stream_profile.name
+        else:
+            stream_profile_name = None
         db.set_model_monitoring_credentials(
             project=self.name,
             credentials={
                 "access_key": access_key,
                 "stream_path": stream_path,
                 "tsdb_connection": tsdb_connection,
+                "tsdb_profile_name": tsdb_profile_name,
+                "stream_profile_name": stream_profile_name,
             },
             replace_creds=replace_creds,
         )

mlrun/secrets.py

@@ -151,7 +151,7 @@ def get_secret_or_env(
     secret_provider: Union[dict, SecretsStore, Callable, None] = None,
     default: Optional[str] = None,
     prefix: Optional[str] = None,
-) -> str:
+) -> Optional[str]:
     """Retrieve value of a secret, either from a user-provided secret store, or from environment variables.
     The function will retrieve a secret value, attempting to find it according to the following order:
 

mlrun/serving/routers.py

@@ -30,7 +30,6 @@ import mlrun.common.model_monitoring
 import mlrun.common.schemas.model_monitoring
 from mlrun.utils import logger, now_date
 
-from .server import GraphServer
 from .utils import RouterToDict, _extract_input_data, _update_result_body
 from .v2_serving import _ModelLogPusher
 
@@ -607,24 +606,24 @@ class VotingEnsemble(ParallelRun):
         self.prediction_col_name = prediction_col_name or "prediction"
         self.format_response_with_col_name_flag = format_response_with_col_name_flag
         self.model_endpoint_uid = None
+        self.model_endpoint = None
         self.shard_by_endpoint = shard_by_endpoint
 
     def post_init(self, mode="sync", **kwargs):
-        server = getattr(self.context, "_server", None) or getattr(
-            self.context, "server", None
-        )
+        server: mlrun.serving.GraphServer = getattr(
+            self.context, "_server", None
+        ) or getattr(self.context, "server", None)
         if not server:
             logger.warn("GraphServer not initialized for VotingEnsemble instance")
             return
-
         if not self.context.is_mock or self.context.monitoring_mock:
-            self.model_endpoint_uid = _init_endpoint_record(
-                server,
-                self,
-                creation_strategy=kwargs.get("creation_strategy"),
-                endpoint_type=kwargs.get("endpoint_type"),
+            self.model_endpoint = mlrun.get_run_db().get_model_endpoint(
+                project=server.project,
+                name=self.name,
+                function_name=server.function_name,
+                function_tag=server.function_tag or "latest",
             )
-
+            self.model_endpoint_uid = self.model_endpoint.metadata.uid
         self._update_weights(self.weights)
 
     def _resolve_route(self, body, urlpath):
@@ -1004,81 +1003,6 @@ class VotingEnsemble(ParallelRun):
                 self._weights[model] = 0
 
 
-def _init_endpoint_record(
-    graph_server: GraphServer,
-    voting_ensemble: VotingEnsemble,
-    creation_strategy: mlrun.common.schemas.ModelEndpointCreationStrategy,
-    endpoint_type: mlrun.common.schemas.EndpointType,
-) -> Union[str, None]:
-    """
-    Initialize model endpoint record and write it into the DB. In general, this method retrieve the unique model
-    endpoint ID which is generated according to the function uri and the model version. If the model endpoint is
-    already exist in the DB, we skip the creation process. Otherwise, it writes the new model endpoint record to the DB.
-
-    :param graph_server:    A GraphServer object which will be used for getting the function uri.
-    :param voting_ensemble: Voting ensemble serving class. It contains important details for the model endpoint record
-                            such as model name, model path, model version, and the ids of the children model endpoints.
-    :param creation_strategy: Strategy for creating or updating the model endpoint:
-        * **overwrite**:
-        1. If model endpoints with the same name exist, delete the `latest` one.
-        2. Create a new model endpoint entry and set it as `latest`.
-        * **inplace** (default):
-        1. If model endpoints with the same name exist, update the `latest` entry.
-        2. Otherwise, create a new entry.
-        * **archive**:
-        1. If model endpoints with the same name exist, preserve them.
-        2. Create a new model endpoint with the same name and set it to `latest`.
-
-    :param endpoint_type:    model endpoint type
-    :return: Model endpoint unique ID.
-    """
-
-    logger.info("Initializing endpoint records")
-    children_uids = []
-    children_names = []
-    for _, c in voting_ensemble.routes.items():
-        if hasattr(c, "endpoint_uid"):
-            children_uids.append(c.endpoint_uid)
-            children_names.append(c.name)
-    try:
-        logger.info(
-            "Creating Or Updating a new model endpoint record",
-            name=voting_ensemble.name,
-            project=graph_server.project,
-            function_name=graph_server.function_name,
-            function_tag=graph_server.function_tag or "latest",
-            model_class=voting_ensemble.__class__.__name__,
-            creation_strategy=creation_strategy,
-        )
-        model_endpoint = mlrun.common.schemas.ModelEndpoint(
-            metadata=mlrun.common.schemas.ModelEndpointMetadata(
-                project=graph_server.project,
-                name=voting_ensemble.name,
-                endpoint_type=endpoint_type,
-            ),
-            spec=mlrun.common.schemas.ModelEndpointSpec(
-                function_name=graph_server.function_name,
-                function_tag=graph_server.function_tag or "latest",
-                model_class=voting_ensemble.__class__.__name__,
-                children_uids=children_uids,
-                children=children_names,
-            ),
-            status=mlrun.common.schemas.ModelEndpointStatus(
-                monitoring_mode=mlrun.common.schemas.model_monitoring.ModelMonitoringMode.enabled
-                if voting_ensemble.context.server.track_models
-                else mlrun.common.schemas.model_monitoring.ModelMonitoringMode.disabled,
-            ),
-        )
-        db = mlrun.get_run_db()
-        db.create_model_endpoint(
-            model_endpoint=model_endpoint, creation_strategy=creation_strategy
-        )
-    except mlrun.errors.MLRunInvalidArgumentError as e:
-        logger.info("Failed to create model endpoint record", error=e)
-        return None
-    return model_endpoint.metadata.uid
-
-
 class EnrichmentModelRouter(ModelRouter):
     """
     Model router with feature enrichment and imputing

mlrun/serving/states.py

@@ -126,6 +126,9 @@ class BaseStep(ModelObj):
         self.shape = shape
         self.on_error = None
         self._on_error_handler = None
+        self.model_endpoint_creation_strategy = (
+            schemas.ModelEndpointCreationStrategy.SKIP
+        )
 
     def get_shape(self):
         """graphviz shape"""
@@ -428,7 +431,7 @@ class TaskStep(BaseStep):
         result_path: Optional[str] = None,
         model_endpoint_creation_strategy: Optional[
             schemas.ModelEndpointCreationStrategy
-        ] = schemas.ModelEndpointCreationStrategy.INPLACE,
+        ] = schemas.ModelEndpointCreationStrategy.SKIP,
         endpoint_type: Optional[schemas.EndpointType] = schemas.EndpointType.NODE_EP,
     ):
         super().__init__(name, after)
@@ -723,6 +726,11 @@ class RouterStep(TaskStep):
         self._routes: ObjectDict = None
         self.routes = routes
         self.endpoint_type = schemas.EndpointType.ROUTER
+        self.model_endpoint_creation_strategy = (
+            schemas.ModelEndpointCreationStrategy.INPLACE
+            if class_name and "serving.VotingEnsemble" in class_name
+            else schemas.ModelEndpointCreationStrategy.SKIP
+        )
 
     def get_children(self):
         """get child steps (routes)"""

mlrun/serving/v2_serving.py

@@ -15,7 +15,7 @@
 import threading
 import time
 import traceback
-from typing import Optional, Union
+from typing import Optional
 
 import mlrun.artifacts
 import mlrun.common.model_monitoring.helpers
@@ -23,7 +23,6 @@ import mlrun.common.schemas.model_monitoring
 import mlrun.model_monitoring
 from mlrun.utils import logger, now_date
 
-from .server import GraphServer
 from .utils import StepToDict, _extract_input_data, _update_result_body
 
 
@@ -114,8 +113,8 @@ class V2ModelServer(StepToDict):
         if model:
             self.model = model
             self.ready = True
-        self._versioned_model_name = None
         self.model_endpoint_uid = None
+        self.model_endpoint = None
         self.shard_by_endpoint = shard_by_endpoint
         self._model_logger = None
 
@@ -139,20 +138,23 @@ class V2ModelServer(StepToDict):
             else:
                 self._load_and_update_state()
 
-        server = getattr(self.context, "_server", None) or getattr(
-            self.context, "server", None
-        )
+        server: mlrun.serving.GraphServer = getattr(
+            self.context, "_server", None
+        ) or getattr(self.context, "server", None)
         if not server:
             logger.warn("GraphServer not initialized for VotingEnsemble instance")
             return
 
+        if not self.context.is_mock and not self.model_spec:
+            self.get_model()
         if not self.context.is_mock or self.context.monitoring_mock:
-            self.model_endpoint_uid = _init_endpoint_record(
-                graph_server=server,
-                model=self,
-                creation_strategy=kwargs.get("creation_strategy"),
-                endpoint_type=kwargs.get("endpoint_type"),
+            self.model_endpoint = mlrun.get_run_db().get_model_endpoint(
+                project=server.project,
+                name=self.name,
+                function_name=server.function_name,
+                function_tag=server.function_tag or "latest",
             )
+            self.model_endpoint_uid = self.model_endpoint.metadata.uid
         self._model_logger = (
             _ModelLogPusher(self, self.context)
             if self.context and self.context.stream.enabled
@@ -233,23 +235,6 @@ class V2ModelServer(StepToDict):
         request = self.preprocess(event_body, op)
         return self.validate(request, op)
 
-    @property
-    def versioned_model_name(self):
-        if self._versioned_model_name:
-            return self._versioned_model_name
-
-        # Generating version model value based on the model name and model version
-        if self.model_path and self.model_path.startswith("store://"):
-            # Enrich the model server with the model artifact metadata
-            self.get_model()
-            if not self.version:
-                # Enrich the model version with the model artifact tag
-                self.version = self.model_spec.tag
-            self.labels = self.model_spec.labels
-        version = self.version or "latest"
-        self._versioned_model_name = f"{self.name}:{version}"
-        return self._versioned_model_name
-
     def do_event(self, event, *args, **kwargs):
         """main model event handler method"""
         start = now_date()
@@ -553,95 +538,3 @@ class _ModelLogPusher:
                 if getattr(self.model, "metrics", None):
                     data["metrics"] = self.model.metrics
                 self.output_stream.push([data], partition_key=partition_key)
-
-
-def _init_endpoint_record(
-    graph_server: GraphServer,
-    model: V2ModelServer,
-    creation_strategy: mlrun.common.schemas.ModelEndpointCreationStrategy,
-    endpoint_type: mlrun.common.schemas.EndpointType,
-) -> Union[str, None]:
-    """
-    Initialize model endpoint record and write it into the DB. In general, this method retrieve the unique model
-    endpoint ID which is generated according to the function uri and the model version. If the model endpoint is
-    already exist in the DB, we skip the creation process. Otherwise, it writes the new model endpoint record to the DB.
-
-    :param graph_server: A GraphServer object which will be used for getting the function uri.
-    :param model:        Base model serving class (v2). It contains important details for the model endpoint record
-                         such as model name, model path, and model version.
-    :param creation_strategy: Strategy for creating or updating the model endpoint:
-        * **overwrite**:
-        1. If model endpoints with the same name exist, delete the `latest` one.
-        2. Create a new model endpoint entry and set it as `latest`.
-        * **inplace** (default):
-        1. If model endpoints with the same name exist, update the `latest` entry.
-        2. Otherwise, create a new entry.
-        * **archive**:
-        1. If model endpoints with the same name exist, preserve them.
-        2. Create a new model endpoint with the same name and set it to `latest`.
-    :param endpoint_type    model endpoint type
-
-    :return: Model endpoint unique ID.
-    """
-
-    logger.info("Initializing endpoint records")
-    if not model.model_spec:
-        model.get_model()
-    if model.model_spec:
-        model_name = model.model_spec.metadata.key
-        model_db_key = model.model_spec.spec.db_key
-        model_uid = model.model_spec.metadata.uid
-        model_tag = model.model_spec.tag
-        model_labels = model.model_spec.labels  # todo : check if we still need this
-    else:
-        model_name = None
-        model_db_key = None
-        model_uid = None
-        model_tag = None
-        model_labels = {}
-    logger.info(
-        "Creating Or Updating a new model endpoint record",
-        name=model.name,
-        project=graph_server.project,
-        function_name=graph_server.function_name,
-        function_tag=graph_server.function_tag or "latest",
-        model_name=model_name,
-        model_tag=model_tag,
-        model_db_key=model_db_key,
-        model_uid=model_uid,
-        model_class=model.__class__.__name__,
-        creation_strategy=creation_strategy,
-        endpoint_type=endpoint_type,
-    )
-    try:
-        model_ep = mlrun.common.schemas.ModelEndpoint(
-            metadata=mlrun.common.schemas.ModelEndpointMetadata(
-                project=graph_server.project,
-                labels=model_labels,
-                name=model.name,
-                endpoint_type=endpoint_type,
-            ),
-            spec=mlrun.common.schemas.ModelEndpointSpec(
-                function_name=graph_server.function_name,
-                function_tag=graph_server.function_tag or "latest",
-                model_name=model_name,
-                model_db_key=model_db_key,
-                model_uid=model_uid,
-                model_class=model.__class__.__name__,
-                model_tag=model_tag,
-            ),
-            status=mlrun.common.schemas.ModelEndpointStatus(
-                monitoring_mode=mlrun.common.schemas.model_monitoring.ModelMonitoringMode.enabled
-                if model.context.server.track_models
-                else mlrun.common.schemas.model_monitoring.ModelMonitoringMode.disabled,
-            ),
-        )
-        db = mlrun.get_run_db()
-        model_ep = db.create_model_endpoint(
-            model_endpoint=model_ep, creation_strategy=creation_strategy
-        )
-    except mlrun.errors.MLRunBadRequestError as e:
-        logger.info("Failed to create model endpoint record", error=e)
-        return None
-
-    return model_ep.metadata.uid

mlrun/utils/version/version.json

@@ -1,4 +1,4 @@
 {
-  "git_commit": "6aa0bfdf77d03774890da6714fc3aca778bb9f26",
-  "version": "1.8.0-rc15"
+  "git_commit": "eb1fcda410abefdd86a344d75af69fe49dc5eb07",
+  "version": "1.8.0-rc17"
 }

{mlrun-1.8.0rc15.dist-info → mlrun-1.8.0rc17.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: mlrun
-Version: 1.8.0rc15
+Version: 1.8.0rc17
 Summary: Tracking and config of machine learning runs
 Home-page: https://github.com/mlrun/mlrun
 Author: Yaron Haviv
@@ -31,7 +31,7 @@ Requires-Dist: ipython~=8.10
 Requires-Dist: nuclio-jupyter~=0.11.1
 Requires-Dist: numpy<1.27.0,>=1.26.4
 Requires-Dist: pandas<2.2,>=1.2
-Requires-Dist: pyarrow<18,>=10.0
+Requires-Dist: pyarrow<17,>=10.0
 Requires-Dist: pyyaml<7,>=6.0.2
 Requires-Dist: requests~=2.32
 Requires-Dist: tabulate~=0.8.6
@@ -88,9 +88,9 @@ Requires-Dist: avro~=1.11; extra == "kafka"
 Provides-Extra: redis
 Requires-Dist: redis~=4.3; extra == "redis"
 Provides-Extra: mlflow
-Requires-Dist: mlflow~=2.8; extra == "mlflow"
+Requires-Dist: mlflow~=2.16; extra == "mlflow"
 Provides-Extra: databricks-sdk
-Requires-Dist: databricks-sdk~=0.13.0; extra == "databricks-sdk"
+Requires-Dist: databricks-sdk~=0.20.0; extra == "databricks-sdk"
 Provides-Extra: sqlalchemy
 Requires-Dist: sqlalchemy~=1.4; extra == "sqlalchemy"
 Provides-Extra: dask
@@ -121,6 +121,7 @@ Requires-Dist: timelength~=1.1; extra == "api"
 Requires-Dist: memray~=1.12; sys_platform != "win32" and extra == "api"
 Requires-Dist: aiosmtplib~=3.0; extra == "api"
 Requires-Dist: pydantic<2,>=1; extra == "api"
+Requires-Dist: mlrun-pipelines-kfp-v1-8[kfp]~=0.3.3; python_version < "3.11" and extra == "api"
 Provides-Extra: all
 Requires-Dist: adlfs==2023.9.0; extra == "all"
 Requires-Dist: aiobotocore<2.16,>=2.5.0; extra == "all"
@@ -131,7 +132,7 @@ Requires-Dist: azure-keyvault-secrets~=4.2; extra == "all"
 Requires-Dist: bokeh>=2.4.2,~=2.4; extra == "all"
 Requires-Dist: boto3<1.36,>=1.28.0; extra == "all"
 Requires-Dist: dask~=2023.12.1; extra == "all"
-Requires-Dist: databricks-sdk~=0.13.0; extra == "all"
+Requires-Dist: databricks-sdk~=0.20.0; extra == "all"
 Requires-Dist: distributed~=2023.12.1; extra == "all"
 Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "all"
 Requires-Dist: google-cloud-bigquery-storage~=2.17; extra == "all"
@@ -140,7 +141,7 @@ Requires-Dist: google-cloud-storage==2.14.0; extra == "all"
 Requires-Dist: google-cloud==0.34; extra == "all"
 Requires-Dist: graphviz~=0.20.0; extra == "all"
 Requires-Dist: kafka-python~=2.0; extra == "all"
-Requires-Dist: mlflow~=2.8; extra == "all"
+Requires-Dist: mlflow~=2.16; extra == "all"
 Requires-Dist: msrest~=0.6.21; extra == "all"
 Requires-Dist: oss2==2.18.1; extra == "all"
 Requires-Dist: ossfs==2023.12.0; extra == "all"
@@ -161,7 +162,7 @@ Requires-Dist: azure-identity~=1.5; extra == "complete"
 Requires-Dist: azure-keyvault-secrets~=4.2; extra == "complete"
 Requires-Dist: boto3<1.36,>=1.28.0; extra == "complete"
 Requires-Dist: dask~=2023.12.1; extra == "complete"
-Requires-Dist: databricks-sdk~=0.13.0; extra == "complete"
+Requires-Dist: databricks-sdk~=0.20.0; extra == "complete"
 Requires-Dist: distributed~=2023.12.1; extra == "complete"
 Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "complete"
 Requires-Dist: google-cloud-bigquery-storage~=2.17; extra == "complete"
@@ -170,7 +171,7 @@ Requires-Dist: google-cloud-storage==2.14.0; extra == "complete"
 Requires-Dist: google-cloud==0.34; extra == "complete"
 Requires-Dist: graphviz~=0.20.0; extra == "complete"
 Requires-Dist: kafka-python~=2.0; extra == "complete"
-Requires-Dist: mlflow~=2.8; extra == "complete"
+Requires-Dist: mlflow~=2.16; extra == "complete"
 Requires-Dist: msrest~=0.6.21; extra == "complete"
 Requires-Dist: oss2==2.18.1; extra == "complete"
 Requires-Dist: ossfs==2023.12.0; extra == "complete"
@@ -195,7 +196,7 @@ Requires-Dist: azure-keyvault-secrets~=4.2; extra == "complete-api"
 Requires-Dist: boto3<1.36,>=1.28.0; extra == "complete-api"
 Requires-Dist: dask-kubernetes~=0.11.0; extra == "complete-api"
 Requires-Dist: dask~=2023.12.1; extra == "complete-api"
-Requires-Dist: databricks-sdk~=0.13.0; extra == "complete-api"
+Requires-Dist: databricks-sdk~=0.20.0; extra == "complete-api"
 Requires-Dist: distributed~=2023.12.1; extra == "complete-api"
 Requires-Dist: fastapi~=0.115.6; extra == "complete-api"
 Requires-Dist: gcsfs<2024.7,>=2023.9.2; extra == "complete-api"
@@ -208,7 +209,8 @@ Requires-Dist: humanfriendly~=10.0; extra == "complete-api"
 Requires-Dist: igz-mgmt~=0.4.1; extra == "complete-api"
 Requires-Dist: kafka-python~=2.0; extra == "complete-api"
 Requires-Dist: memray~=1.12; sys_platform != "win32" and extra == "complete-api"
-Requires-Dist: mlflow~=2.8; extra == "complete-api"
+Requires-Dist: mlflow~=2.16; extra == "complete-api"
+Requires-Dist: mlrun-pipelines-kfp-v1-8[kfp]~=0.3.3; python_version < "3.11" and extra == "complete-api"
 Requires-Dist: msrest~=0.6.21; extra == "complete-api"
 Requires-Dist: objgraph~=3.6; extra == "complete-api"
 Requires-Dist: oss2==2.18.1; extra == "complete-api"