How to run an Azure ML Model

Question

I need help to run my Azure ML Model for my lasso pattern detector project. I have created the model, but now I'm not sure how to input data and run it to receive an output. Additionally, I cannot create a Real-time endpoint, and I don't have access to App Registration. Can someone provide me with a step-by-step guide, preferably starting from the very beginning? My data is in a CSV file with a column named "label" and other columns named "Pixels" for each part of the pattern (1 and 0), and I can't use Image classification because I have no quota.

Answer 1

Hi @Alaxcia,

Thank you for reaching out to Microsoft Q&A forum!

Here's a step-by-step guide to run the Azure ML model:

• Create an Azure Machine Learning workspace by following this quickstart.
• Create an Azure Machine Learning experiment: In your workspace, create a new experiment.
• Upload your CSV file by clicking on the "Data" tab and then clicking "Data assets" tab and create new. Select "From Local File" and upload your CSV file. Make sure to specify the column names and data types.
• Create an Azure Machine Learning model by this documentation.
• Once you have created the model, you can deploy it as a web service. And after deploying the model, you can test it by sending a request to the web service.

Regarding your issue with creating a Real-time endpoint and not having access to App Registration, you can follow the instructions in this Azure documentation to create a Real-time endpoint. If you don't have access or permission to App Registration, you can ask your Azure AD administrator to assign the role to you. If you are the Azure AD administrator, you can follow the instructions in this Azure documentation to create a custom role with the necessary permissions for App Registration.

Regarding quota related issues see here: Manage quotas and limits.

For more info see complete azure ML documentation.

I hope this helps! Thank you.

---

Please do not forget to click Accept Answer and Yes for was this answer helpful, wherever the information provided helps you. This can be beneficial to other community members.