This sample shows how to leverage ADF to run an experiment.
Create Azure Machine Learning Preview accounts and install Azure Machine Learning Workbench using these instructions.
First, create compute targets for running the parameter sweep: DSVM and optionally, HDInsight Spark Cluster. Select File, Open Command Prompt and enter following commands to create the compute targets.
Create a Ubuntu baesd DSVM using these instructions. Attach it as compute target, and then prepare it by using:
$ az ml computetarget attach --name <dsvm> --address <dsvm-ip> --username <sshusername> --password <sshpwd> --type remotedocker
$ az ml experiment prepare -c <dsvm>
OR
Create HDInsight Spark Cluster using these instructions. Attach it as compute target, and then prepare it by using:
$ az ml computetarget attach --name <myhdi> --address <myhdi-ssh.azurehdinsight.net> --username <sshusername> --password <sshpwd> --type cluster
$ az ml experiment prepare -c <myhdi>
DSVM running a pyspark job.
$ az ml experiment submit -c <dsvm> <pyspark.py>
HDInsight Spark running a pyspark job.
$ az ml experiment submit -c <myhdi> <pyspark.py>
The creation process creates a new service princple account when run locally. This new account must be used to login to the az ml client from within the workbench if the AMLW secret store is used to store passwords.
- Download and place IaC folder within your AML Workspace project directory.
- Download and place batch task within your AML Workspace project directory.
- Select File, Open Powershell and enter the following command to create the deployment.
- Make VSTS Git Access Key
- DSVM Compute Target
- Path in AML Workbench project to code
.\IaC\CreateDeployment.ps1 `
-gitPassword "<vstsPat>" `
-dsvm "<dsvm>" `
-pythonPath "<pyspark.py>"
RDP into Node using username rdpuser, and provided password.
- microsoft web platform installer is on the desktop of DSVM, use this to update Powershell
- Download AML Workbench from here, https://aka.ms/azureml-wb-msi
- Windows Key
- Type cmd
- Right click, select Run as Admin
- Change to D drive with d:
- cd User{username}\Downloads
- AmlWorkbenchSetup.msi
- Log into AML Workbench, this is required to create starter folders, remote experiment submission will fail if this is not complete.
This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.microsoft.com.
When you submit a pull request, a CLA-bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., label, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.
This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.
©2017 Microsoft Corporation. All rights reserved. This information is provided "as-is" and may change without notice. Microsoft makes no warranties, express or implied, with respect to the information provided here.