NotebookRunnerStep 类
创建在 Azure 机器学习中将本地笔记本作为管道步骤运行的步骤。
- 继承
-
azureml.pipeline.core._python_script_step_base._PythonScriptStepBaseNotebookRunnerStep
构造函数
NotebookRunnerStep(name=None, notebook_run_config=None, runconfig_pipeline_params=None, compute_target=None, params=None, inputs=None, outputs=None, allow_reuse=True, version=None, output_notebook_pipeline_data_name=None)
参数
- runconfig_pipeline_params
- dict[str, PipelineParameter]
默认值: None
在运行时使用键值对替代 runconfig 属性,每个键值对分别具有 runconfig 属性的名称和该属性的 PipelineParameter。
- compute_target
- Union[<xref:azureml.core.compute.DsvmCompute,azureml.core.compute.AmlCompute,azureml.core.compute.RemoteCompute,str>]
默认值: None
[必需] 要使用的计算目标。
- params
- dict
默认值: None
可在笔记本中访问的名称-值对字典。 这是 azureml.contrib.notebook.NotebookRunConfig 中的参数的附加项 可在此处提供 azureml.pipeline.core.PipelineParameters。
- inputs
- list[Union[<xref:azureml.pipeline.core.graph.InputPortBinding,azureml.data.data_reference.DataReference,azureml.pipeline.core.PortDataReference,azureml.pipeline.core.builder.PipelineData,azureml.data.dataset_consumption_config.DatasetConsumptionConfig,azureml.pipeline.core.pipeline_output_dataset.PipelineOutputFileDataset>]]
默认值: None
输入端口绑定的列表。
- outputs
- list[Union[<xref:azureml.pipeline.core.builder.PipelineData,azureml.pipeline.core.graph.OutputPortBinding,azureml.pipeline.core.pipeline_output_dataset.PipelineOutputFileDataset>]]
默认值: None
输出端口绑定的列表。
- allow_reuse
- bool
默认值: True
指示当使用相同的设置重新运行时,该步骤是否应重用以前的结果。 默认情况下启用重用。 如果步骤内容(脚本/依赖项)以及输入和参数保持不变,则重用此步骤以前运行的输出。 重用该步骤时,不是将作业提交到计算,而是使前一运行的结果立即可供后续步骤使用。
- output_notebook_pipeline_data_name
- str
默认值: None
包含此运行的输出笔记本(其中包含每个单元格的输出)的中间管道数据的名称。 指定此项会生成一个输出笔记本(作为此步骤的输出之一),它可以在管道中进一步进行传递。
方法
| create_node |
为 Python 脚本步骤创建节点。 |
create_node
为 Python 脚本步骤创建节点。
create_node(graph, default_datastore, context)
参数
- context
- <xref:azureml.pipeline.core._GraphContext>
必需
图上下文。
返回
已创建的节点。
返回类型
反馈
提交和查看相关反馈