question

23174178 avatar image
1 Vote"
23174178 asked ramr-msft commented

How can I save my machine learning prediciton model which is built in Azure ML Studio?

I want to extract or save the model which I worked and built in Azure ML Studio as .pkl file or some other format. Can anyone please share the steps on how to do it?

azure-machine-learning-studio-classic
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

1 Answer

ramr-msft avatar image
0 Votes"
ramr-msft answered ramr-msft commented

@23174178 Thanks for reaching out. Microsoft link to Download the entry script file and conda dependencies file which explains how we can get access to score.py and conda_env.yaml files under Output + logs tab for Train module.
The document shared is enabling deploy designer trained model in general AML deploying way. You need to register your model at first (the step before you download conda and score.py), and then use the conda file and score.py to deploy the model. However deploying in this way does not include data processing part in your training pipeline.


For real-time inference:
If you want to deploy a real-time endpoint which includes data preprocessing, deploying from real-time inference pipeline is recommended, and you can follow this tutorial.


· 2
5 |1600 characters needed characters left characters exceeded

Up to 10 attachments (including images) can be used with a maximum of 3.0 MiB each and 30.0 MiB total.

@ramr-msft Thanks for the information.
But I am not able get into the Window where the document is addressing to !

Currently I am in the ML Studio (Classic). This is my student account (1 year free subscription with credits of 100 points).

Attached is for your reference.

If possible can you please show how to go in the respective window & save the model through GIF file.

0 Votes 0 ·

@23174178 Thanks for the details. We recommend using the latest Azure machine learning Designer instead of the ML Studio (Classic).


0 Votes 0 ·