frogml-core 0.0.1__py3-none-any.whl → 0.0.2__py3-none-any.whl

frogml_core/frogml_client/client.py

@@ -0,0 +1,876 @@
+from datetime import timedelta
+from functools import lru_cache
+from pathlib import Path
+from tempfile import TemporaryDirectory
+from typing import TYPE_CHECKING, Dict, List, Optional, Union
+
+from frogml_proto.qwak.build.v1.build_pb2 import DESCRIPTOR
+from frogml_core.clients.analytics.client import AnalyticsEngineClient
+from frogml_core.clients.automation_management.client import AutomationsManagementClient
+from frogml_core.clients.batch_job_management.client import BatchJobManagerClient
+from frogml_core.clients.build_management.client import BuildsManagementClient
+from frogml_core.clients.build_orchestrator.client import (
+    BuildFilter,
+    BuildOrchestratorClient,
+)
+from frogml_core.clients.data_versioning.client import DataVersioningManagementClient
+from frogml_core.clients.data_versioning.data_tag_filter import DataTagFilter
+from frogml_core.clients.deployment.client import DeploymentManagementClient
+from frogml_core.clients.feature_store import FeatureRegistryClient
+from frogml_core.clients.file_versioning.client import FileVersioningManagementClient
+from frogml_core.clients.file_versioning.file_tag_filter import FileTagFilter
+from frogml_core.clients.instance_template.client import (
+    InstanceTemplateManagementClient,
+)
+from frogml_core.clients.jfrog_gateway.client import JfrogGatewayClient
+from frogml_core.clients.model_management.client import ModelsManagementClient
+from frogml_core.clients.model_version_manager import ModelVersionManagerClient
+from frogml_core.clients.project.client import ProjectsManagementClient
+from frogml_core.exceptions import FrogmlException
+from frogml_core.feature_store.feature_sets.batch import BatchFeatureSet
+from frogml_core.frogml_client.batch_jobs.execution import Execution
+from frogml_core.frogml_client.batch_jobs.task import Task
+from frogml_core.frogml_client.builds.build import Build
+from frogml_core.frogml_client.builds.filters.metric_filter import MetricFilter
+from frogml_core.frogml_client.builds.filters.parameter_filter import ParameterFilter
+from frogml_core.frogml_client.data_versioning.data_tag import DataTag
+from frogml_core.frogml_client.deployments.deployment import (
+    Deployment,
+    EnvironmentAudienceRoute,
+)
+from frogml_core.frogml_client.file_versioning.file_tag import FileTag
+from frogml_core.frogml_client.models.model import Model
+from frogml_core.frogml_client.models.model_metadata import ModelMetadata
+from frogml_core.frogml_client.projects.project import Project
+from frogml_core.inner.build_config.build_config_v1 import BuildConfigV1
+from frogml_core.inner.tool.auth import FrogMLAuthClient
+from frogml_core.model.base import BaseModel
+
+if TYPE_CHECKING:
+    try:
+        import pandas as pd
+    except ImportError:
+        pass
+
+
+class FrogMLClient:
+    """
+    Comprehensive Frogml client to manage operations performed against the Frogml system.
+    Acts as a wrapper and single entry point for the different Frogml clients - BatchClient,
+    DeploymentClient, and more.
+    """
+
+    @lru_cache(maxsize=1)
+    def _get_project_management(self) -> ProjectsManagementClient:
+        return ProjectsManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_model_management(self) -> ModelsManagementClient:
+        return ModelsManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_build_management(self) -> BuildsManagementClient:
+        return BuildsManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_build_orchestrator(self) -> BuildOrchestratorClient:
+        return BuildOrchestratorClient()
+
+    @lru_cache(maxsize=1)
+    def _get_batch_job_manager(self) -> BatchJobManagerClient:
+        return BatchJobManagerClient()
+
+    @lru_cache(maxsize=1)
+    def _get_file_versioning_management(self) -> FileVersioningManagementClient:
+        return FileVersioningManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_data_versioning_management(self) -> DataVersioningManagementClient:
+        return DataVersioningManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_deployment_management(self) -> DeploymentManagementClient:
+        return DeploymentManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_feature_registry_client(self) -> FeatureRegistryClient:
+        return FeatureRegistryClient()
+
+    @lru_cache(maxsize=1)
+    def _get_automations_client(self) -> AutomationsManagementClient:
+        return AutomationsManagementClient()
+
+    @lru_cache(maxsize=1)
+    def _get_analytics_engine(self) -> AnalyticsEngineClient:
+        return AnalyticsEngineClient()
+
+    @lru_cache(maxsize=1)
+    def _get_instance_template_client(self) -> InstanceTemplateManagementClient:
+        return InstanceTemplateManagementClient()
+
+    @staticmethod
+    @lru_cache(maxsize=1)
+    def _get_auth_client():
+        return FrogMLAuthClient()
+
+    @lru_cache(maxsize=1)
+    def _get_model_version_manager_client(self) -> ModelVersionManagerClient:
+        return ModelVersionManagerClient()
+
+    @lru_cache(maxsize=1)
+    def _get_jfrog_gateway_client(self) -> JfrogGatewayClient:
+        return JfrogGatewayClient()
+
+    def get_token(self) -> str:
+        """
+        Retrieve a token by api key.
+        Use Cases:
+        Default - from already configured (frogml configured) - conf file
+        Support FROGML_API_KEY environment variable
+        Get token for a specified API key
+        Returns:
+            str: a client token
+        """
+        client = self._get_auth_client()
+        return client.get_token()
+
+    def get_latest_build(
+        self, model_id: str, build_status: str = "SUCCESSFUL"
+    ) -> Optional[str]:
+        """
+        Get the latest build by its model ID.
+        Optionally gets a build_status, by default filters on 'SUCCESSFUL'
+
+        Args:
+            model_id (str): The model ID
+            build_status (str): build statuses to filter on. Valid values are 'SUCCESSFUL', 'IN_PROGRESS', 'FAILED'
+
+        Returns:
+             str: The build ID of the latest build according to the build status. None if no builds match the filter.
+
+        """
+        if (
+            build_status
+            not in DESCRIPTOR.enum_types_by_name["BuildStatus"].values_by_name
+        ):
+            raise FrogmlException(
+                f"Invalid build status {build_status}. Valid options are 'SUCCESSFUL', 'IN_PROGRESS', 'FAILED'"
+            )
+
+        model = self._get_model_management().get_model(model_id=model_id)
+        model_uuid = model.uuid
+
+        builds = self._get_build_orchestrator().list_builds(model_uuid=model_uuid).build
+        if not builds:
+            return None
+
+        builds = [
+            build
+            for build in builds
+            if build.build_status
+            == DESCRIPTOR.enum_types_by_name["BuildStatus"]
+            .values_by_name[build_status]
+            .number
+        ]
+        builds.sort(key=lambda r: r.audit.created_at.seconds)
+
+        return builds[-1].buildId if builds else None
+
+    def get_builds_by_tags(
+        self,
+        model_id: str,
+        tags: List[str],
+        match_any: bool = True,
+        include_extra_tags: bool = True,
+    ) -> List[Build]:
+        """
+        Get builds by its model ID and explicit tags
+
+        Args:
+            model_id (str): The model ID
+            tags (List[str]): List of tags to filter by
+            match_any (bool): Whether matching any tag is enough to return a build (default: True)
+            include_extra_tags (bool): Whether builds are allowed to have more tags than the ones specified in the tags list (default: True)
+
+        Returns:
+             List[Build]: List of builds that contains the requested tags.
+
+        """
+        model = self._get_model_management().get_model(model_id=model_id)
+        model_uuid = model.uuid
+
+        builds_proto = (
+            self._get_build_orchestrator()
+            .list_builds(
+                model_uuid=model_uuid,
+                build_filter=BuildFilter(
+                    tags=tags,
+                    require_all_tags=not match_any,
+                    include_extra_tags=include_extra_tags,
+                ),
+            )
+            .build
+        )
+
+        builds = [Build.from_proto(build) for build in builds_proto]
+
+        return builds if builds else None
+
+    def get_build(self, build_id: str) -> Build:
+        """
+        Get builds by its build ID
+
+        Args:
+            build_id (str): The build ID
+
+        Returns:
+             Build: The requested build.
+
+        """
+        build_proto = self._get_build_orchestrator().get_build(build_id).build
+
+        build = Build.from_proto(build_proto)
+
+        if build:
+            model_proto = self._get_model_management().get_model_by_uuid(
+                model_uuid=build_proto.model_uuid
+            )
+            build.model_id = model_proto.model_id
+
+        return build if build else None
+
+    def list_builds(
+        self,
+        model_id: str,
+        tags: List[str],
+        filters: List[Union[MetricFilter, ParameterFilter]],
+    ) -> List[Build]:
+        """
+        List builds by its model ID and explicit filters
+
+        Args:
+            model_id (str): The model ID
+            tags (List[str]): List of tags to filter by
+            filters (List[str]): List of metric and parameter filters
+
+        Returns:
+             List[Build]: List of builds that contains the requested filters.
+
+        """
+        model = self._get_model_management().get_model(model_id=model_id)
+        model_uuid = model.uuid
+
+        metric_filters = list()
+        parameter_filters = list()
+
+        for build_filter in filters:
+            if isinstance(build_filter, MetricFilter):
+                metric_filters.append(MetricFilter.to_proto(build_filter))
+            else:
+                parameter_filters.append(ParameterFilter.to_proto(build_filter))
+
+        builds_proto = (
+            self._get_build_orchestrator()
+            .list_builds(
+                model_uuid=model_uuid,
+                build_filter=BuildFilter(
+                    tags=tags,
+                    metric_filters=metric_filters,
+                    parameter_filters=parameter_filters,
+                ),
+            )
+            .build
+        )
+
+        return [Build.from_proto(build) for build in builds_proto]
+
+    def list_file_tags(
+        self,
+        model_id: str,
+        build_id: str = "",
+        filter: Optional[FileTagFilter] = None,
+    ) -> List[FileTag]:
+        """
+        List file tags by its model ID
+
+        Args:
+            model_id (str): The model ID
+            build_id (str): The build ID - optional.
+                If not specified, returns all model file tags.
+            filter (FileTagFilter): Filter list by - optional.
+                value (str): The filter value.
+                type (enum): The filter type - FILE_TAG_FILTER_TYPE_CONTAINS/FILE_TAG_FILTER_TYPE_PREFIX.
+                If not specified, returns all model file tags.
+
+
+        Returns:
+             List[FileTag]: List of file tags with their specifications.
+
+        """
+        model_file_tags_proto = (
+            self._get_file_versioning_management().get_model_file_tags(
+                model_id=model_id,
+                build_id=build_id,
+                file_tag_filter=None if filter is None else filter.to_proto(),
+            )
+        )
+
+        return [
+            FileTag.from_proto(file_tag) for file_tag in model_file_tags_proto.file_tags
+        ]
+
+    def list_data_tags(
+        self,
+        model_id: str,
+        build_id: str = "",
+        filter: Optional[DataTagFilter] = None,
+    ) -> List[DataTag]:
+        """
+        List data tags by its model ID
+
+        Args:
+            model_id (str): The model ID
+            build_id (str): The build ID - optional.
+                If not specified, returns all model data tags.
+            filter (DataTagFilter): Filter list by - optional.
+                value (str): The filter value.
+                type (enum): The filter type - DATA_TAG_FILTER_TYPE_CONTAINS/DATA_TAG_FILTER_TYPE_PREFIX.
+                If not specified, returns all model file tags.
+
+        Returns:
+             List[DataTag]: List of data tags with their specifications.
+
+        """
+        model_data_tags_proto = (
+            self._get_data_versioning_management().get_model_data_tags(
+                model_id=model_id,
+                build_id=build_id,
+                data_tag_filter=None if filter is None else filter.to_proto(),
+            )
+        )
+
+        return [
+            DataTag.from_proto(data_tag) for data_tag in model_data_tags_proto.data_tags
+        ]
+
+    def set_tag(self, build_id: str, tag: str) -> None:
+        """
+        Assign a tag to an existing build
+
+        Args:
+            build_id (str): The build ID
+            tag (str): The tag to assign
+
+        """
+        self._get_build_orchestrator().register_tags(build_id=build_id, tags=[tag])
+
+    def set_tags(self, build_id: str, tags: List[str]) -> None:
+        """
+        Assign a list of tags to an existing build
+
+        Args:
+            build_id (str): The build ID
+            tags (List[str]): List of tags to assign
+        """
+        self._get_build_orchestrator().register_tags(build_id=build_id, tags=tags)
+
+    def create_project(self, project_name: str, project_description: str) -> str:
+        """
+        Create project
+
+        Args:
+            project_name (str): The requested name
+            project_description (str): The requested description
+
+        Returns:
+             str: The project ID of the newly created project
+
+        """
+        project = self._get_project_management().create_project(
+            project_name=project_name, project_description=project_description
+        )
+
+        return project.project.project_id
+
+    def get_project(self, project_id: str) -> Optional[Project]:
+        """
+        Get model by its project ID
+
+        Args:
+            project_id (str): The model ID
+
+        Returns:
+             Optional[Project]: Project by ID.
+
+        """
+        return Project.from_proto(
+            self._get_project_management().get_project(project_id).project.spec
+        )
+
+    def list_projects(self) -> List[Project]:
+        """
+        List projects
+
+        Returns:
+             List[Project]: List of projects
+
+        """
+        projects_proto = self._get_project_management().list_projects().projects
+
+        return [Project.from_proto(project) for project in projects_proto]
+
+    def delete_project(self, project_id: str) -> None:
+        """
+        Delete project by its project ID
+
+        Args:
+            project_id (str): The project ID
+        """
+        return self._get_project_management().delete_project(project_id=project_id)
+
+    def create_model(
+        self, project_id: str, model_name: str, model_description: str
+    ) -> str:
+        """
+        Create model
+
+        Args:
+            project_id (str): The project ID to associate the model
+            model_name (str): The requested name
+            model_description (str): The requested description
+
+        Returns:
+             str: The model ID of the newly created project
+
+        """
+        model = self._get_model_management().create_model(
+            project_id=project_id,
+            model_name=model_name,
+            model_description=model_description,
+        )
+
+        return model.model_id
+
+    def get_model(self, model_id: str) -> Optional[Model]:
+        """
+        Get model by its model ID
+
+        Args:
+            model_id (str): The model ID
+
+        Returns:
+             Optional[BaseModel]: Model by ID.
+
+        """
+        return BaseModel.from_proto(
+            self._get_model_management().get_model(model_id=model_id)
+        )
+
+    def list_model_metadata(self, project_id: str) -> List[ModelMetadata]:
+        """
+        Lists metadata of all models in a project.
+        Args:
+            project_id (str): The project ID
+        Returns:
+            List[ModelMetadata]: List of model metadata
+        """
+        service_response = self._get_model_management().list_models_metadata(project_id)
+        model_metadata_from_service = service_response.model_metadata
+
+        result = []
+        for metadata_source in model_metadata_from_service:
+            model_metadata = ModelMetadata()
+
+            model_metadata.model = BaseModel.from_proto(metadata_source.model)
+            model_metadata.deployed_builds = [
+                Build.from_proto(build) for build in metadata_source.build
+            ]
+            model_metadata.deployments = [
+                Deployment.from_proto(deployment)
+                for deployment in metadata_source.deployment_details
+            ]
+
+            audience_routes = []
+            for (
+                env_id,
+                env,
+            ) in metadata_source.audience_routes_grouped_by_environment.items():
+                for route in env.audience_routes:
+                    audience_routes.append(
+                        EnvironmentAudienceRoute.from_proto(route, env_id)
+                    )
+
+            model_metadata.audience_routes = audience_routes
+            result.append(model_metadata)
+
+        return result
+
+    def list_models(self, project_id: str) -> List[Model]:
+        """
+        List models
+
+        Args:
+            project_id (str): The project ID to list models from
+
+        Returns:
+             List[BaseModel]: List of models
+
+        """
+        models_proto = (
+            self._get_model_management().list_models(project_id=project_id).models
+        )
+
+        return [Model.from_proto(model) for model in models_proto]
+
+    def get_model_metadata(self, model_id: str) -> ModelMetadata:
+        """
+        Get model metadata by its model ID
+
+        Args:
+            model_id (str): The model ID
+
+        Returns:
+            Model metadata by ID.
+        """
+        service_response = self._get_model_management().get_model_metadata(model_id)
+        model_metadata_from_service = service_response.model_metadata
+        model_metadata = ModelMetadata()
+
+        model_metadata.model = Model.from_proto(model_metadata_from_service.model)
+        model_metadata.deployed_builds = [
+            Build.from_builds_management(build)
+            for build in model_metadata_from_service.build
+        ]
+        model_metadata.deployments = [
+            Deployment.from_proto(deployment)
+            for deployment in model_metadata_from_service.deployment_details
+        ]
+
+        audience_routes = []
+        for (
+            env_id,
+            env,
+        ) in model_metadata_from_service.audience_routes_grouped_by_environment.items():
+            for route in env.audience_routes:
+                audience_routes.append(
+                    EnvironmentAudienceRoute.from_proto(env_id, route)
+                )
+        model_metadata.audience_routes = audience_routes
+
+        return model_metadata
+
+    def list_models_metadata(self, project_id: str) -> List[ModelMetadata]:
+        """
+        List models metadata
+
+        Args:
+            project_id (str): The project ID to list models metadata from
+
+        Returns:
+             List[ModelMetadata]: List of models metadata
+
+        """
+        models_metadata_proto = (
+            self._get_model_management()
+            .list_models_metadata(project_id=project_id)
+            .model_metadata
+        )
+
+        return [
+            ModelMetadata.from_proto(model_metadata)
+            for model_metadata in models_metadata_proto
+        ]
+
+    def delete_model(self, model_id: str) -> None:
+        """
+        Delete model by its model ID
+
+        Args:
+            model_id (str): The model ID
+        """
+        return self._get_model_management().delete_model(
+            project_id=self.get_model(model_id).project_id, model_id=model_id
+        )
+
+    def get_deployed_build_id_per_environment(
+        self, model_id: str
+    ) -> Optional[Dict[str, Dict[str, str]]]:
+        """
+        Get deployed build ID per environment by its model ID
+
+        Args:
+            model_id (str): The model ID
+
+        Returns:
+             Dict[str, str]: Map environment to deployed build ID. None if the model is not deployed.
+
+        """
+        model = self._get_model_management().get_model(model_id=model_id)
+        model_uuid = model.uuid
+
+        deployment_details = self._get_deployment_management().get_deployment_details(
+            model_id=model_id, model_uuid=model_uuid
+        )
+
+        if not deployment_details:
+            return None
+
+        environment_to_deployment_details = (
+            deployment_details.environment_to_deployment_details
+        )
+
+        return {
+            env_id: {
+                deployment.variation.name: deployment.build_id
+                for deployment in v.deployments_details
+            }
+            for env_id, v in environment_to_deployment_details.items()
+        }
+
+    def run_analytics_query(
+        self, query: str, timeout: timedelta = None
+    ) -> "pd.DataFrame":
+        """
+        Runs a Frogml Analytics Query and returns the Pandas DataFrame with the results.
+
+        Args:
+            query (str): The query to run
+            timeout (timedelta): The timeout for the query - optional.
+                If not specified, the function will wait indefinitely.
+
+        Returns:
+            pd.DataFrame: The results of the query
+        """
+        try:
+            import pandas as pd
+        except ImportError:
+            raise FrogmlException(
+                "Missing Pandas dependency required for running an analytics query."
+            )
+        client = self._get_analytics_engine()
+        url = client.get_analytics_data(query=query, timeout=timeout)
+        return pd.read_csv(url)
+
+    def list_feature_sets(self) -> List[BatchFeatureSet]:
+        """
+        List all feature sets
+
+        Returns:
+            List[FeatureSet]: List of feature sets
+        """
+        client = self._get_feature_registry_client()
+        feature_set_protos = client.list_feature_sets().feature_families
+
+        return [
+            BatchFeatureSet._from_proto(
+                feature_set.feature_set_definition.feature_set_spec
+            )
+            for feature_set in feature_set_protos
+        ]
+
+    def delete_feature_set(self, feature_set_name: str):
+        """
+        Delete a feature set
+
+        Args:
+            feature_set_name (str): The feature set name
+        """
+        client = self._get_feature_registry_client()
+        feature_set_def = client.get_feature_set_by_name(
+            feature_set_name=feature_set_name
+        )
+        if feature_set_def:
+            client.delete_feature_set(
+                feature_set_def.feature_set.feature_set_definition.feature_set_id
+            )
+        else:
+            raise FrogmlException(f"Feature set named '{feature_set_name}' not found")
+
+    def delete_featureset_version(self, featureset_name: str, version_number: int):
+        """
+        Delete a feature set version
+
+        Args:
+            featureset_name (str): The feature set name
+            version_number (int): version number
+        """
+        client = self._get_feature_registry_client()
+
+        client.delete_featureset_version(
+            featureset_name=featureset_name, version_number=version_number
+        )
+
+    def set_featureset_active_version(self, featureset_name: str, version_number: int):
+        """
+        Set active feature set version
+
+        Args:
+            featureset_name (str): The feature set name
+            version_number (int): version number
+        """
+        client = self._get_feature_registry_client()
+
+        client.set_active_featureset_version(
+            featureset_name=featureset_name, version_number=version_number
+        )
+
+    def delete_data_source(self, data_source_name: str):
+        """
+        Delete a data source
+
+        Args:
+            data_source_name (str): The data source name
+        """
+        client = self._get_feature_registry_client()
+        data_source_def = client.get_data_source_by_name(
+            data_source_name=data_source_name
+        )
+        if data_source_def:
+            client.delete_data_source(
+                data_source_def.data_source.data_source_definition.data_source_id
+            )
+        else:
+            raise FrogmlException(f"Data source named '{data_source_name}' not found")
+
+    def delete_entity(self, entity_name: str):
+        """
+        Delete an entity_name
+
+        Args:
+            entity_name (str): The entity name
+        """
+        client = self._get_feature_registry_client()
+        entity_def = client.get_entity_by_name(entity_name=entity_name)
+        if entity_def:
+            client.delete_entity(entity_def.entity.entity_definition.entity_id)
+        else:
+            raise FrogmlException(f"Entity named '{entity_name}' not found")
+
+    def trigger_batch_feature_set(self, feature_set_name):
+        """
+        Triggers a batch feature set ingestion job immediately.
+
+        Args:
+            feature_set_name (str): The feature set name
+        """
+        client = self._get_feature_registry_client()
+        client.run_feature_set(feature_set_name=feature_set_name)
+
+    def trigger_automation(self, automation_name):
+        """
+        Triggers an automation immediately.
+
+        Args:
+            automation_name (str): The automation name
+        """
+        client = self._get_automations_client()
+        automation = client.get_automation_by_name(automation_name=automation_name)
+        if not automation:
+            raise FrogmlException(
+                f"Could not find automation with given name '{automation_name}'"
+            )
+        client.run_automation(automation_id=automation.id)
+
+    def build_model(
+        self,
+        model_id: str,
+        main_module_path: str = ".",
+        dependencies_file: Optional[str] = None,
+        dependencies_list: Optional[List[str]] = None,
+        tags: Optional[List[str]] = None,
+        instance: str = "small",
+        gpu_compatible: bool = False,
+        run_tests: bool = True,
+        validate_build_artifact: bool = True,
+        validate_build_artifact_timeout: Optional[int] = None,
+        prebuilt_frogml_model: Optional[Model] = None,
+        base_image: Optional[str] = None,
+        build_name: str = "",
+    ):
+        from frogml_core.frogml_client.build_api_helpers.trigger_build_api import (
+            trigger_build_api,
+        )
+
+        config = BuildConfigV1()
+        config.build_properties.model_id = model_id
+        config.build_properties.build_name = build_name
+        config.build_properties.model_uri.uri = main_module_path
+        config.build_env.python_env.dependency_file_path = dependencies_file
+        config.build_properties.tags = tags if tags else []
+        config.build_env.remote.resources.instance = instance
+        config.build_properties.gpu_compatible = gpu_compatible
+        config.step.validate_build_artifact = validate_build_artifact
+        config.step.tests = run_tests
+        config.fetch_base_docker_image_name(
+            self._get_build_orchestrator(), self._get_instance_template_client()
+        )
+        config.pre_built_model = prebuilt_frogml_model
+        if validate_build_artifact_timeout:
+            config.step.validate_build_artifact_timeout = (
+                validate_build_artifact_timeout
+            )
+        if base_image:
+            config.build_env.docker.base_image = base_image
+        if dependencies_list:
+            with TemporaryDirectory() as temporary_directory:
+                requirements_filename = Path(temporary_directory) / "requirements.txt"
+                with open(requirements_filename, "w") as temp_dependencies_file:
+                    temp_dependencies_file.write("\n".join(dependencies_list))
+                config.build_env.python_env.dependency_file_path = requirements_filename
+                return trigger_build_api(config)
+        else:
+            return trigger_build_api(config)
+
+    def list_executions(
+        self,
+        model_id: str,
+        build_id: str = "",
+    ) -> List[Execution]:
+        """
+        List batch executions by its model ID
+
+        Args:
+            model_id (str): The model ID
+            build_id (str): The build ID - optional.
+                If not specified, returns all model executions.
+
+
+        Returns:
+             List[Execution]: List of executions with their specifications.
+
+        """
+        model_executions_proto = self._get_batch_job_manager().list_batch_jobs(
+            model_id=model_id,
+            build_id=build_id,
+        )
+
+        return [
+            Execution.from_proto(execution)
+            for execution in model_executions_proto.batch_jobs
+        ]
+
+    def list_execution_tasks(
+        self,
+        execution_id: str,
+    ) -> List[Task]:
+        """
+        List batch executions tasks by its job ID
+
+        Args:
+            execution_id (str): The execution ID
+
+
+        Returns:
+             List[Task]: List of execution tasks with their specifications.
+
+        """
+        model_execution_tasks_proto = (
+            self._get_batch_job_manager().get_batch_job_details(
+                job_id=execution_id,
+            )
+        )
+
+        return [
+            Task.from_proto(task)
+            for task in model_execution_tasks_proto.batch_job.task_executions
+        ]