entry_point_loaders 模块

包含将 Azure 机器学习与 MLFlow 集成的加载程序。

函数

azureml_artifacts_builder

为 AzureMLflow 创建项目存储库。

azureml_artifacts_builder(artifact_uri=None)

参数

artifact_uri
str
默认值: None

存储项目的 URI。

azureml_model_registry_builder

通过 MLflow 在 Azure 中为模型创建或返回注册表。

azureml_model_registry_builder(store_uri)

参数

store_uri
str
必需

注册表的 URI。

azureml_project_run_builder

为 MLflow 项目创建一个空的 AzureMLProjectBackend 对象以访问运行功能。

azureml_project_run_builder()

azureml_request_header_provider

通过 MLflow 在 Azure 中创建或返回请求标头提供程序。

azureml_request_header_provider()

azureml_store_builder

创建或返回存储以通过 MLflow 在 Azure 中读取和记录指标和项目。

azureml_store_builder(store_uri, artifact_uri=None)

参数

store_uri
str
必需

存储的 URI。

artifact_uri
str
默认值: None

存储项目的 URI。