{"id":17885,"date":"2018-06-05T21:36:25","date_gmt":"2018-06-06T04:36:25","guid":{"rendered":"https:\/\/blogs.msdn.microsoft.com\/dotnet\/?p=17885"},"modified":"2021-09-29T16:25:18","modified_gmt":"2021-09-29T23:25:18","slug":"announcing-ml-net-0-2","status":"publish","type":"post","link":"https:\/\/devblogs.microsoft.com\/dotnet\/announcing-ml-net-0-2\/","title":{"rendered":"Announcing ML.NET 0.2"},"content":{"rendered":"

Last month at \/\/Build 2018, we released ML.NET 0.1, a cross-platform, open source machine learning framework<\/a>. We would like to thank the community for the engagement so far in helping us shape ML.NET.<\/p>\n

Today we are releasing ML.NET 0.2. This release focuses on adding new ML tasks like clustering, making it easier to validate models, adding a brand-new repo for ML.NET samples and addressing a variety of issues and feedback we received in the GitHub repo<\/a>.<\/p>\n

Some of the highlights with ML.NET 0.2 release are mentioned below.<\/p>\n

New Machine Learning Tasks: Clustering<\/h3>\n

Clustering is an unsupervised learning task that groups sets of items based on their features. It identifies which items are more similar to each other than other items.<\/p>\n

This might be useful in scenarios such as organizing news articles into groups based on their topics, segmenting users based on their shopping habits, and grouping viewers based on their taste in movies.<\/p>\n

The Iris Flower sample<\/a> illustrates how you can use Clustering with ML.NET 0.2<\/p>\n

Easier model validation with cross-validation and train-test<\/h3>\n

Cross-validation<\/a> is an approach to validating how well your model statistically performs. It does not require a separate test dataset, but rather uses your training data to test your model (it partitions the data so different data is used for training and testing, and it does this multiple times).With ML.NET 0.2 you can now use cross-validation and here is a good example<\/a>.<\/p>\n

Train-test<\/strong> is a shortcut to testing your model on a separate dataset. See example usage here.<\/a><\/span><\/p>\n

Train using data objects with CollectionDataSource<\/h3>\n

ML.NET 0.1 enabled loading data from a delimited text file.\u00a0CollectionDataSource\u00a0in ML.NET 0.2 adds the ability to use a collection of objects as the input to a\u00a0LearningPipeline.\n\n<\/strong>The code-snippet below shows how you can use CollectionDataSource with ML.NET 0.2.<\/p>\n

var pipeline = new LearningPipeline();\r\nvar data = new List() {\r\n           new IrisData { SepalLength = 1f, SepalWidth = 1f \r\n                         ,PetalLength=0.3f, PetalWidth=5.1f, Label=1},\r\n           new IrisData { SepalLength = 1f, SepalWidth = 1f \r\n                         ,PetalLength=0.3f, PetalWidth=5.1f, Label=1},\r\n           new IrisData { SepalLength = 1.2f, SepalWidth = 0.5f \r\n                         ,PetalLength=0.3f, PetalWidth=5.1f, Label=0}\r\n           };\r\nvar collection = CollectionDataSource.Create(data);\r\npipeline.Add(collection); <\/code><\/pre>\n
Full code snippet for CollectionDataSource can be found here<\/a>.<\/p>\n
New ML.NET samples repo<\/h3>\n
We have created a new repo https:\/\/github.com\/dotnet\/machinelearning-samples<\/a> and added a few getting started and end-end app samples.<\/p>\n