Flowthru.Integrations.MLNet
0.1.6
See the version list below for details.
dotnet add package Flowthru.Integrations.MLNet --version 0.1.6
NuGet\Install-Package Flowthru.Integrations.MLNet -Version 0.1.6
<PackageReference Include="Flowthru.Integrations.MLNet" Version="0.1.6" />
<PackageVersion Include="Flowthru.Integrations.MLNet" Version="0.1.6" />
<PackageReference Include="Flowthru.Integrations.MLNet" />
paket add Flowthru.Integrations.MLNet --version 0.1.6
#r "nuget: Flowthru.Integrations.MLNet, 0.1.6"
#:package Flowthru.Integrations.MLNet@0.1.6
#addin nuget:?package=Flowthru.Integrations.MLNet&version=0.1.6
#tool nuget:?package=Flowthru.Integrations.MLNet&version=0.1.6
Flowthru.Integrations.MLNet
ML.NET integration for Flowthru - provides IDataView container adapters for machine learning pipelines.
Purpose
This integration library enables seamless integration between Flowthru's data catalog system and ML.NET's machine learning framework. It provides container adapters that bridge Flowthru's streaming row abstraction with ML.NET's columnar IDataView representation.
Key Components
DataViewContainerAdapter<T>
Container adapter that converts between IAsyncEnumerable<T> rows and ML.NET's IDataView columnar format.
Features:
- Native ML.NET integration
- Type-safe row schema
- Lazy evaluation support
- Optimized for ML.NET pipelines
Usage:
using Flowthru.Integrations.MLNet.Container;
using Microsoft.ML;
public record FeatureRow(
float Feature1,
float Feature2,
int Label
) : IFlatSchema, ITextSerializable;
var mlContext = new MLContext();
var adapter = new DataViewContainerAdapter<FeatureRow>(mlContext);
// Load data from Flowthru catalog
var dataView = await adapter.FromRows(rowStream);
// Use with ML.NET pipelines
var pipeline = mlContext.Transforms
.NormalizeMinMax("Feature1")
.Append(mlContext.Transforms.NormalizeMinMax("Feature2"));
var model = pipeline.Fit(dataView);
Dependencies
- Flowthru - Core data catalog and storage abstractions
- Microsoft.ML - ML.NET framework
- LanguageExt.Core - Functional programming primitives
Integration with ML.Next
This adapter bridges Flowthru catalogs with ML.Next's type-safe wrappers, enabling end-to-end compile-time safety for ML pipelines.
Architecture
This is an optional integration - the core Flowthru library does not depend on ML.NET. Users who need ML.NET support can add this package as a separate dependency.
Flowthru (core)
↑
│ depends on
│
Flowthru.Integrations.MLNet (optional)
↑
│ depends on
│
Microsoft.ML
Version
0.1.0 (Pre-alpha)
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- Flowthru (>= 0.1.6)
- LanguageExt.Core (>= 5.0.0-beta-54)
- Microsoft.ML (>= 4.0.0)
- Microsoft.ML.OnnxRuntime (>= 1.20.1)
- Microsoft.ML.OnnxTransformer (>= 4.0.0)
NuGet packages
This package is not used by any NuGet packages.
GitHub repositories
This package is not used by any popular GitHub repositories.