Search code examples
pythonazureazure-machine-learning-service

Select columns dynamically in Azure ML model

I have deployed a model as a Webservice in Azure ML.Its a simple one and all it does is do a linear Regression .The underlying code is python . Now i need to pass which all columns have to selected as independent variables, dynamically, from the client side . How may i do this in Azure ML studio?

Solution

  • Based on my understanding, I think you want to dynamically get the selected columns data via request the Azure ML webservice with some parameters on the client.

    You can refer to the offical document Use Azure Machine Learning Web Service Parameters and the blog AzureML Web Service Parameters to know how to set and use the web service parameters to implement your needs via add the selected column names as array into the json parameter GlobalParameters.

    Meanwhile, there is a client sample on GitHub https://github.com/nk773/AzureML_RRSApp. Althought it was writen in Java, I think it is easy to understand, then you can rewrite in Python with requests package.