search

Azure / MachineLearningNotebooks

Python notebooks with ML and deep learning examples with Azure Machine Learning Python SDK | Microsoft
https://docs.microsoft.com/azure/machine-learning/service/
MIT License
4.07k stars 2.52k forks source link

RunConfiguration example #555

Closed nansravn closed 4 years ago

nansravn commented 5 years ago

We have the following statement on the Ramarks section: "see the python code example in the documentation for RunConfiguration for an example of a ScriptRunConfig object in action."

I've searched on the RunConfiguration doc, but couldn't find any example of a ScriptRunConfig object in action.

Could you add an example about how to use ScriptRunConfig?

I'm reviewing some MLOps samples, I have an Azure DevOps hosted environment that's already configured with all the requirements for running successfully my experiment. I want to submit a training code from this DevOps hosted env, to be execute locally on the same environment. I don't want to specify additional conda_env packages or nothing like that.

Do I only need to use RunConfiguration(), with default parameters, on my ScriptRunConfig()?

Document Details

⚠ Do not edit this section. It is required for docs.microsoft.com ➟ GitHub issue linking.

vmagelo commented 4 years ago

We added a sample. Look for it in the next release of the SDK in 2 weeks. Thanks!

vmagelo commented 4 years ago

Here's something that should get you going:

from azureml.core import ScriptRunConfig, RunConfiguration, Experiment

# create or load an experiment experiment = Experiment(workspace, "MyExperiment") # run a trial from the train.py code in your current directory config = ScriptRunConfig(source_directory='.', script='train.py', run_config=RunConfiguration()) script_run = experiment.submit(config)

Also take a look at:

j-martens commented 4 years ago

Thanks for your report. We hope this answers your question. We will now proceed to close this issue.