Machine Learning with ML.Net
Download as PPTX, PDF0 likes390 views
The document discusses ML.NET, a machine learning framework for .NET. It provides an overview of machine learning concepts and types. It then describes ML.NET's capabilities like using existing .NET skills for custom models, supporting frameworks like TensorFlow, and automated machine learning tools like the ML.NET Model Builder visual studio extension and ML.NET CLI. These tools make it easy for developers to build, train, and deploy machine learning models without extensive ML experience.
Machine Learning with ML.Net
- 1. www.Suhail.Cloud #SuhailCloud @SuhailCloud Machine Learning with ML.NET Suhail Jamaldeen
- 2. www.Suhail.Cloud #SuhailCloud @SuhailCloud About Me Suhail Jamaldeen Consultant & Trainer – Office 365 | Azure Microsoft MVP – Office Apps & Services Speaker Blogger – www.Suhail.Cloud #SuhailCloud
- 3. www.Suhail.Cloud #SuhailCloud @SuhailCloud Agenda .Net Core Machine Learning ML.Net Automated Machine Learning ML.Net Model Builder ML.Net CLI
- 4. www.Suhail.Cloud #SuhailCloud @SuhailCloud .NET 5 = .NET Core vNext 3.0 Platform to build anything
- 5. www.Suhail.Cloud #SuhailCloud @SuhailCloud Machine Learning • Machine learning is a subset of Artificial Intelligence (AI) • Provides systems the ability to automatically learn and improve from experience without being explicitly programmed. • Process • Prepare Data • Build Model (Acquire Knowledge) • Use Model (Predict Decisions) • Results are probabilistic
- 6. www.Suhail.Cloud #SuhailCloud @SuhailCloud Machine Learning Workflow
- 7. www.Suhail.Cloud #SuhailCloud @SuhailCloud Machine Learning Types Supervised Learning Classification • Binary classification – Sentiment analysis • Multi classification – Issue Classification Regression • Price prediction Unsupervised Learning Clustering • Cluster customers into who purchase similar products
- 8. www.Suhail.Cloud #SuhailCloud @SuhailCloud Artificial Intelligence vs. Machine Learning vs. Deep Learning
- 9. www.Suhail.Cloud #SuhailCloud @SuhailCloud ML.Net Machine Learning framework for building custom ML Models Cross-platform and open-source available Announced
- 10. www.Suhail.Cloud #SuhailCloud @SuhailCloud Reuse .Net Skills C# and F# for custom models Proven at scale Azure, Windows, Office Used in PowerBI, Outlook, Visual Studio… Extensible Tenserflow, ONNX and Infer.Net Custom LM made easy Automated ML and Tools (Model builder and CLI)
- 11. www.Suhail.Cloud #SuhailCloud @SuhailCloud Few samples from GitHub repository https://github.com/dotnet/machinelearning-samples
- 15. www.Suhail.Cloud #SuhailCloud @SuhailCloud ML.Net Machine Learning Samples https://github.com/dotnet/machinelearning- samples
- 16. www.Suhail.Cloud #SuhailCloud @SuhailCloud Automated Machine Learning – AutoML (Preview state) Automate the creation of the model • You don't need to write the code by yourself to train a model • You simply need to provide your datasets. • The "best" model and the code for running it will be generated for you. Currently only support Binary-Classification, Multiclass Classification and Regression. In upcoming versions will be supporting additional ML Tasks such as Recommendations, Anomaly Detection, Clustering, etc..
- 17. www.Suhail.Cloud #SuhailCloud @SuhailCloud Available in three form factors • ML.Net Model Builder • ML.Net CLI • ML.Net AutoML API (For ISVs)
- 18. www.Suhail.Cloud #SuhailCloud @SuhailCloud ML.Net Model Builder (Preview) • UI tooling in Visual Studio as Visual studio extension • Uses Automated Machine Learning (AutoML) to easily allow to build, train and ship custom machine learning models • Developers with no ML expertise can use this tool easily • Model Builder currently supports Regression, Binary-Classification and Multi-Classification tasks.
- 19. www.Suhail.Cloud #SuhailCloud @SuhailCloud • Currently support .tsv, .csv, .txt, and SQL as the data types • If you have a .txt file, columns should be separated with ',' or ';' or '/t‘ • The files must have a header row.
- 20. www.Suhail.Cloud #SuhailCloud @SuhailCloud ML.Net Model Builder www.aka.ms/dotnetmodelbuilder https://github.com/dotnet/machinelearning- samples/tree/master/modelbuilder
- 21. www.Suhail.Cloud #SuhailCloud @SuhailCloud ML.Net CLI (Preview state) The ML.NET CLI is a tool you can run on any command-prompt • Windows – Poweshell and CMD • Mac or Linux – Bash Generates good quality ML.NET models based on training datasets you provide. It also generates sample C# code to run/score the model Generates C# code that was used to create/train it so you can research what algorithm and settings it is using.
- 23. www.Suhail.Cloud #SuhailCloud @SuhailCloud Currently, the ML Tasks supported by the ML.NET CLI are: • binary-classification • multiclass-classification • regression Future: other machine learning tasks such as recommendation, ranking, anomaly-detection, clustering dotnet tool install -g mlnet mlnet auto-train --task binary-classification --dataset "wikipedia-detox-250-line- data.tsv" --label-column-name Sentiment --max-exploration-time 30
- 24. www.Suhail.Cloud #SuhailCloud @SuhailCloud ML.Net CLI https://docs.microsoft.com/en- us/dotnet/machine-learning/automate- training-with-cli
- 25. www.Suhail.Cloud #SuhailCloud @SuhailCloud Machine Learning Workflow
- 26. www.Suhail.Cloud #SuhailCloud @SuhailCloud References https://dotnet.microsoft.com/ https://github.com/dotnet/machinelearning-samples https://dotnet.microsoft.com/learn/ml-dotnet
- 28. www.Suhail.Cloud #SuhailCloud @SuhailCloud Thank You www.Suhail.Cloud @SuhailCloud #SuhailCloud
Editor's Notes
- #9: We will be targeting Supervised Learning is the target
- #10: Runs everywhere
- #11: ONNX = Open Neural Network Exchange (ONNX)
- #15: Show the samples fro the audiences https://github.com/dotnet/machinelearning-samples https://blazorsentimentanalysisproduction.azurewebsites.net/
- #18: ISVs = independent software vendor
- #19: https://github.com/dotnet/machinelearning-samples/tree/master/modelbuilder
- #22: CLI (command-line interface)