machine-learning-sample-experiments.md

title	description	services	documentationcenter	author	manager	editor	ms.assetid	ms.service	ms.workload	ms.tgt_pltfrm	ms.devlang	ms.topic	ms.date	ms.author
Copy machine learning sample experiments | Microsoft Docs	Learn how to use sample machine learning experiments to create new experiments with Cortana Intelligence Gallery and Microsoft Azure Machine Learning.	machine-learning		cjgronlund	jhubbard	cgronlun	81e6c1d8-682c-4db3-bfd5-d7bfb1150ff3	machine-learning	data-services	na	na	get-started-article	11/11/2016	cgronlun;olgali

Copy sample experiments to create new machine learning experiments

Learn how to start with sample experiments from Cortana Intelligence Gallery instead of creating machine learning experiments from scratch. You can use the samples to build your own machine learning solution.

In the gallery are sample experiments by the Microsoft Azure Machine Learning team as well as samples shared by the Machine Learning community. You also can ask questions or post comments about experiments.

To see how to use the gallery, watch the 3-minute video Copy other people's work to do data science from the series Data Science for Beginners.

[!INCLUDE machine-learning-free-trial]

Find an experiment to copy in Cortana Intelligence Gallery

To see what experiments are available, go to the Gallery and click Experiments at the top of the page.

Find the newest or most popular experiments

On this page, you can see Recently added experiments, or scroll down to look at What's popular or the latest Popular Microsoft experiments.

Look for an experiment that meets specific requirements

To browse all experiments:

1. Click Browse all at the top of the page.

2. Under Refine by, select Experiment to see all the experiments in the Gallery.

3. You can find experiments that meet your requirements a couple different ways:

   • Select filters on the left. For example, to browse experiments that use a PCA-based anomaly detection algorithm, select Experiment under Categories, and PCA-Based Anomaly Detection under Algorithms Used. (If you don't see that algorithm, click Show all at the bottom of the list.)
     
     
   • Use the search box. For example, to find experiments contributed by Microsoft related to digit recognition that use a two-class support vector machine algorithm, enter "digit recognition" in the search box. Then, select the filters Experiment, Microsoft content only, and Two-Class Support Vector Machine:

4. Click an experiment to learn more about it.

5. To run and/or modify the experiment, click Open in Studio on the experiment's page.

   [!NOTE] To open an experiment in Machine Learning Studio, you need to sign in with your Microsoft account credentials. If you don’t have a Machine Learning workspace yet, a free trial workspace is created. Learn what’s included in the Machine Learning free trial

   

Use a template in Machine Learning Studio

You also can create a new experiment in Machine Learning Studio using a Gallery sample as a template.

1. Sign in with your Microsoft account credentials to the Studio, and then click New to create a new experiment.
2. Browse through the sample content and click one.

A new experiment is created in your workspace using the sample experiment as a template.

Next steps

• Prepare your data
• Try using R in your experiment
• Review sample R experiments
• Create a web service API