我如何在ML Azure Pipeline中使用环境

Question

背景

我已经从conda environment.yml以及一些docker配置和环境变量创建了ML Workspace环境。我可以在Python笔记本中访问它：

env = Environment.get(workspace=ws, name='my-environment', version='1')

我可以成功地使用它来运行Python脚本作为实验，即

runconfig = ScriptRunConfig(source_directory='script/', script='my-script.py', arguments=script_params)
runconfig.run_config.target = compute_target
runconfig.run_config.environment = env
run = exp.submit(runconfig)

问题

我现在想与管道运行相同的脚本，以便可以使用不同的参数触发多次运行。我创建了管道，如下所示：

pipeline_step = PythonScriptStep(
    source_directory='script', script_name='my-script.py',
    arguments=['-a', param1, '-b', param2],
    compute_target=compute_target,
    runconfig=runconfig
)
steps = [pipeline_step]
pipeline = Pipeline(workspace=ws, steps=steps)
pipeline.validate()

然后尝试运行管道时：

pipeline_run = Experiment(ws, 'my_pipeline_run').submit(
    pipeline, pipeline_parameters={...}
)

我收到以下错误：Response status code does not indicate success: 400 (Conda dependencies were not specified. Please make sure that all conda dependencies were specified i).

[当我查看在Azure门户中运行的管道时，似乎尚未拾取环境：没有配置我的conda依赖项，因此代码无法运行。我在做什么错？

Answer 1

您快到了，但是您需要使用RunConfiguration而不是ScriptRunConfig。更多信息here

from azureml.core.runconfig import RunConfiguration

env = Environment.get(workspace=ws, name='my-environment', version='1')
# create a new runconfig object
runconfig = RunConfiguration()
runconfig.environment = env

pipeline_step = PythonScriptStep(
    source_directory='script', script_name='my-script.py',
    arguments=['-a', param1, '-b', param2],
    compute_target=compute_target,
    runconfig=runconfig
)

pipeline = Pipeline(workspace=ws, steps=[pipeline_step])

pipeline_run = Experiment(ws, 'my_pipeline_run').submit(pipeline)