microsoft/dstoolkit-classification-solution-accelerator

This is a classification solution accelerator to help you build and deploy a binary classification project.

dstoolkit-classification-solution-accelerator's Issues

github markdown syntax about images & TOC

There seem to be some rendering issues about images like EndToEndSetup

Sample Jupyter notebook

We don't have any Notebooks in /notebooks folder, and can we have any sample notebooks with the function defined in /src scripts?

import `Workspace` in Azure ML

I found that there're some options about the setting of Workspace in Azure ML in /src/utils.py script, and they're ordered sequentially. Indeed, they consist of 3 steps with try - except structure:

1. Induce from run object

    try:
        run = Run.get_context()
        if not isinstance(run, _OfflineRun):
            ws = run.experiment.workspace
            return ws
    except Exception as ex:
        print('Workspace from run not found', ex)

2. Retrieve config file with .from_config() method

    try:
        ws = Workspace.from_config()
        return ws
    except Exception as ex:
        print('Workspace config not found in local folder', ex)

3. Pre-defined subscription & Service principal

    try:
        sp = ServicePrincipalAuthentication(
            tenant_id=os.environ['AML_TENANT_ID'],
            service_principal_id=os.environ['AML_PRINCIPAL_ID'],
            service_principal_password=os.environ['AML_PRINCIPAL_PASS']
        )
        ws = Workspace.get(
            name="<ml-example>",
            auth=sp,
            subscription_id="<your-sub-id>"
        )
    except Exception as ex:
        print('Workspace config not found in project', ex)

Imagine that an user wants to use option 3(Service principal), and he/she receives errors for both options 1/2. Then, is it better to modify the structure of try - except?