Machine Learning
para devs com
ML.NET
@LetticiaNicoli
Olá!
Eu sou a Letticia Nicoli
o Software Engineer no @Nubank
o Microsoft MVP – Xamarin, .NET
o Organizadora do @High5Devs
o Cursando MBA em AI & ML
@LetticiaNicoli
Machine Learning
O que é essa “buzzword”?
“
Campo de estudo que dá aos computadores a
habilidade de aprender sem serem
explicitamente programados.
Datasets
ML model ML model
App/tools for training
the ML model
End-user app
using the ML
model
Prepare your data Build & Train Run
Model Creation Model Consumption
Machine Learning Workflow
ML.NET
■ Um framework para machine learning open source e
cross-platform.
http://dot.net/ml
Built for
.NET developers
Custom ML made
easy with tools
Extended with
TensorFlow & more
Trusted &
proven at scale
C#
F#
DESKTOP CLOUDWEB MOBILE ML
.NET
Your platform for building anything
IoTGAMING
§ Frameworks suportados:
○ .NET Core (Natively)
○ .NET Framework (Natively)
○ Python with NimbusML (Python bindings)
ML.NET runs anywhere
A few things you can do with ML.NET …
And more! Samples @ https://github.com/dotnet/machinelearning-samples
Formas de utilizar o ML.NET
C#
ML.NET
API
(Code)
ML.NET
Model Builder
(Visual Studio UI)
>_
ML.NET
CLI
(Command-Line
Interface)
ML.NET Model Builder
• Interface simples para criação
de modelos com AutoML;
• Carregamento de arquivos e
banco de dados;
• Geração de código para
treinamento e consumo.
http://aka.ms/mlnetmodelbuilder
AutoML: Classificação binária, multi-classe e regressão
ML.NET CLI
• Criação de modelos com AutoML;
• Cross platform
(Windows, Linux, MacOS);
• Geração de código para
treinamento e consumo.
> mlnet auto-train --ml-task binary-classification --dataset
"customer-reviews.tsv" --label-column-name Sentiment
macOS / Linux (Bash)
Windows (PowerShell e CMD)
Demo
Análise de Sentimentos
ML.NET CLI
Alta performance e acurácia
Microsoft Defender – Antivirus Threat Protection
Power Point - Design Ideas
Excel - Chart Recommendations
Bing - Ad Predictions
Azure ML Studio – Multiple components
Azure Stream Analytics - Anomaly Detection
Brenmor – Medical patient survey classification
Sig Parser – E-mail spam detection
Williams Mullen – Law document classification
Evolution Software – Hazelnut drying time
prediction
endjin – Newsletter article classification
Microsoft Products
(Using ML.NET internally for > 8 years)
Customers
(ML.NET v1 since May 2019)
Outras features importantes no ML.NET
• DevOps & ML Model lifecycle with Azure DevOps CI/CD pipelines
• Model Explainability and Feature input variables importance
• Cross-validation of ML.NET models:
• Loading data from ‘anywhere’ through the LoadFromEnumerable() method
• Using huge and sparse datasets (thousands or millions of columns)
• Deploy ML.NET model into high scalable and multi-threaded ASP.NET Core apps and services
• Deploy ML.NET model into an Azure Function
• Plus many more ML Tasks and scenarios (see samples here: https://github.com/dotnet/machinelearning-samples )
such as: Sentiment analysis, Spam detection, Fraud Detection, text classification, products recommendations, data
spike detection, clustering, ranking results, etc.
https://aka.ms/mlnetyoutube
Check out many of those scenarios at the ML.NET YouTube playlist:
ML.NET Resources
http://dot.net/ml
http://aka.ms/mlnetsamples
http://aka.ms/mlnetdocs
http://aka.ms/mlnet
https://aka.ms/mlnetyoutube
Obrigada!
Dúvidas?
@LetticiaNicoli
letticia.nicoli@gmail.com

Machine Learning para devs com ML.NET

  • 1.
    Machine Learning para devscom ML.NET @LetticiaNicoli
  • 2.
    Olá! Eu sou aLetticia Nicoli o Software Engineer no @Nubank o Microsoft MVP – Xamarin, .NET o Organizadora do @High5Devs o Cursando MBA em AI & ML @LetticiaNicoli
  • 3.
    Machine Learning O queé essa “buzzword”?
  • 4.
    “ Campo de estudoque dá aos computadores a habilidade de aprender sem serem explicitamente programados.
  • 6.
    Datasets ML model MLmodel App/tools for training the ML model End-user app using the ML model Prepare your data Build & Train Run Model Creation Model Consumption Machine Learning Workflow
  • 7.
    ML.NET ■ Um frameworkpara machine learning open source e cross-platform. http://dot.net/ml Built for .NET developers Custom ML made easy with tools Extended with TensorFlow & more Trusted & proven at scale C# F#
  • 8.
    DESKTOP CLOUDWEB MOBILEML .NET Your platform for building anything IoTGAMING
  • 9.
    § Frameworks suportados: ○.NET Core (Natively) ○ .NET Framework (Natively) ○ Python with NimbusML (Python bindings) ML.NET runs anywhere
  • 10.
    A few thingsyou can do with ML.NET … And more! Samples @ https://github.com/dotnet/machinelearning-samples
  • 11.
    Formas de utilizaro ML.NET C# ML.NET API (Code) ML.NET Model Builder (Visual Studio UI) >_ ML.NET CLI (Command-Line Interface)
  • 12.
    ML.NET Model Builder •Interface simples para criação de modelos com AutoML; • Carregamento de arquivos e banco de dados; • Geração de código para treinamento e consumo. http://aka.ms/mlnetmodelbuilder AutoML: Classificação binária, multi-classe e regressão
  • 13.
    ML.NET CLI • Criaçãode modelos com AutoML; • Cross platform (Windows, Linux, MacOS); • Geração de código para treinamento e consumo. > mlnet auto-train --ml-task binary-classification --dataset "customer-reviews.tsv" --label-column-name Sentiment macOS / Linux (Bash) Windows (PowerShell e CMD)
  • 14.
  • 15.
  • 16.
    Microsoft Defender –Antivirus Threat Protection Power Point - Design Ideas Excel - Chart Recommendations Bing - Ad Predictions Azure ML Studio – Multiple components Azure Stream Analytics - Anomaly Detection Brenmor – Medical patient survey classification Sig Parser – E-mail spam detection Williams Mullen – Law document classification Evolution Software – Hazelnut drying time prediction endjin – Newsletter article classification Microsoft Products (Using ML.NET internally for > 8 years) Customers (ML.NET v1 since May 2019)
  • 17.
    Outras features importantesno ML.NET • DevOps & ML Model lifecycle with Azure DevOps CI/CD pipelines • Model Explainability and Feature input variables importance • Cross-validation of ML.NET models: • Loading data from ‘anywhere’ through the LoadFromEnumerable() method • Using huge and sparse datasets (thousands or millions of columns) • Deploy ML.NET model into high scalable and multi-threaded ASP.NET Core apps and services • Deploy ML.NET model into an Azure Function • Plus many more ML Tasks and scenarios (see samples here: https://github.com/dotnet/machinelearning-samples ) such as: Sentiment analysis, Spam detection, Fraud Detection, text classification, products recommendations, data spike detection, clustering, ranking results, etc. https://aka.ms/mlnetyoutube Check out many of those scenarios at the ML.NET YouTube playlist:
  • 18.
  • 19.