Share via

Azure AI: ONNX model in Automated ML

sara rezaei 0 Reputation points
2024-05-15T11:17:37.0066667+00:00

I am currently exploring the capabilities of Azure Machine Learning Studio for running automated machine learning (Automated ML) experiments. Specifically, I am interested in enabling ONNX compatible models for regression tasks within the Automated ML.

However, after reviewing the Azure Machine Learning Studio UI, I couldn't find an option to enable ONNX compatible models directly through the interface when configuring AutoML experiments.

Could you please provide information regarding the availability of this feature in the Azure Machine Learning Studio UI for AutoML experiments? If it is currently not available through the UI, are there any alternative methods or workarounds to enable ONNX compatible models for regression tasks when running Automated ML experiments?

Azure Machine Learning
Azure Machine Learning
An Azure machine learning service for building and deploying models.
2,630 questions

1 answer

Sort by: Most helpful
Most helpful Newest Oldest
  1. dupammi 7,575 Reputation points Microsoft Vendor
    2024-05-16T09:33:37.4233333+00:00

    Hi @sara rezaei

    Thank you for your question.

    Currently, the Azure Machine Learning Studio UI does not have an option to enable ONNX compatible models directly when configuring AutoML experiments. However, you can download ONNX model files from an AutoML training run using the Azure Machine Learning Python SDK or the Azure Machine Learning studio UI.

    To download ONNX model files from the Azure Machine Learning studio UI, go to your experiment, select the best child run, and then go to Outputs+logs > train_artifacts. You can then manually download the labels.json and model.onnx files.

    Alternatively, you can use the Azure Machine Learning Python SDK to download the labels.json and model.onnx files. You can select the best child run (by primary metric) with the experiment name and parent run ID. Then, you can download the labels.json and model.onnx files using the SDK.

    Once you have the ONNX model files, you can use ONNX Runtime for Python to perform inference on the model. You can find more related information on how to use ONNX Runtime for Python in the Azure documentation. SDK Samples are available for ONNX here, but I don't think there are any options for the UI.

    I hope this helps. Thank you.

    Please don't forget to click Accept Answer and Yes for was this answer helpful.

    0 comments