How to train a model using Azure ML

How to train a model using Azure ML

1) Launching a terminal for the compute instance

click on compute instance and then "..." near jupyter notebook and launch terminal (if still not launched, use "+" symbol to launch the terminal)

2) installing azure-ai-ml from the terminal

pip uninstall azure-ai-ml

>>> by default none installed

pip install azure-ai-ml

Actual output:

>>> copy/paste wasnt possible (it finished fine with exception for blog and file-share with compatible warnings)

3) cloning azure-ml-labs git repo

git clone https://github.com/MicrosoftLearning/mslearn-azure-ml.git azure-ml-labs

4)  Verify we get new folders created

Users/your-user-name/azure-ml-labs folder has been created.

>> yes we have

5) Open the Labs/02/"Run training script.ipynb" notebook.

6) Verify that the notebook uses the Python 3.8 - AzureML kernel. Each kernel has its own image with its own set of packages pre-installed.

7) Run all cells in the notebook.

>>> here we needed to reauthenticate the visual studio, post reauthentication, we can continue running the cell. The job should run and complete.

8) Check the job progress

Queued -> Preparing -> Running -> finalizing -> Completed

Once job is complete, we can click on the job to see the model's accuracy.