8. ML.NET is for building custom models
Custom models
Easier / Less Control Harder / Full Control
Pre-built models
TensorFlow
ML.NETVisionSpeech LanguageKnowledge Search
9. Easy / Less Control Full Control / Harder
Vision Speech Language
Knowledge SearchLabs
TextAnalyticsAPI client = new TextAnalyticsAPI();
client.AzureRegion = AzureRegions.Westus;
client.SubscriptionKey = "1bf33391DeadFish";
client.Sentiment(
new MultiLanguageBatchInput(
new List<MultiLanguageInput>()
{
new MultiLanguageInput("en","0",
"This vacuum cleaner sucks so much dirt")
}));
e.g. Sentiment Analysis using Azure Cognitive Services
9% positive
Pre-built ML Models (Azure Cognitive Services)
10. Easy / Less Control
Full Control / Harder
Prepare Your Data Build & Train Run
Build your own (custom) ML Models
11. Less Control / Easy
Existing Solutions
Build your own (custom) ML Models
12. 02 WHAT IS ML.NET?
ML.NET is a
framework first
14. 03 WHAT IS ML.NET?
ML.NET has been
proven in large scale
Microsoft products
15. ML.NET Usage at Microsoft
+ more!
Windows 10
Power Point
Excel
Bing Ads
16. 04 WHAT IS ML.NET?
ML.NET is
Open Source
& Cross-Platform
17. Microsoft Confidential
Proven & Extensible Open Source
https://github.com/dotnet/machinelearning
Build your own
Supported on Windows, Linux, and macOS
Developer Focused
ML.NET 0.4.0 (Preview)
Machine Learning framework made for .NET developers
21. Load Data Extract Features Train Model Evaluate Model Model consumption
labels + plain text labels + feature vectors model
22. Load Data Extract Features Train Model Evaluate Model Model consumption
labels + plain text labels + feature vectors
Enter...
in ML.NETLearningPipelines!
model
23. Load Data Extract Features Train Model Evaluate Model Model consumption