EricLouchez.Shorokoo.Core
0.1.7-dev
dotnet add package EricLouchez.Shorokoo.Core --version 0.1.7-dev
NuGet\Install-Package EricLouchez.Shorokoo.Core -Version 0.1.7-dev
<PackageReference Include="EricLouchez.Shorokoo.Core" Version="0.1.7-dev" />
<PackageVersion Include="EricLouchez.Shorokoo.Core" Version="0.1.7-dev" />
<PackageReference Include="EricLouchez.Shorokoo.Core" />
paket add EricLouchez.Shorokoo.Core --version 0.1.7-dev
#r "nuget: EricLouchez.Shorokoo.Core, 0.1.7-dev"
#:package EricLouchez.Shorokoo.Core@0.1.7-dev
#addin nuget:?package=EricLouchez.Shorokoo.Core&version=0.1.7-dev&prerelease
#tool nuget:?package=EricLouchez.Shorokoo.Core&version=0.1.7-dev&prerelease
Shorokoo
Define, train, and run neural networks in pure C# — no Python required.
- Define models as C# classes — strongly typed
Tensor<float32>,Scalar<int64>, shapes checked as you build. - Train them — reverse-mode autodiff, optimizers, learning-rate schedules, checkpointing.
- Run them fast — execution backed by ONNX Runtime on CPU or GPU.
- Interoperate — export models as
.onnx, load pretrained weights from.safetensors.
Getting started
Install this package plus one backend for your platform, and (recommended)
the source generator for the [Module] syntax:
dotnet add package Shorokoo
dotnet add package Shorokoo.LinuxCPU # or Shorokoo.LinuxGPU / Shorokoo.WinCPU / Shorokoo.WinGPU
dotnet add package Shorokoo.CodeGen
For ready-made layers, losses, and optimizers also add:
dotnet add package Shorokoo.Modules
using Shorokoo;
using Shorokoo.Modules;
using static Shorokoo.Globals;
using static Shorokoo.NN;
[Module]
public partial class Dense
{
public static Tensor<float32> Inline(
[Hyper] Scalar<int64> outFeatures,
Tensor<float32> x)
{
// ... build the layer from tensor ops ...
}
}
Documentation
Guides, API reference, and samples: https://github.com/Eric-Louchez/Shorokoo
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net8.0 is compatible. net8.0-android was computed. net8.0-browser was computed. net8.0-ios was computed. net8.0-maccatalyst was computed. net8.0-macos was computed. net8.0-tvos was computed. net8.0-windows was computed. net9.0 was computed. net9.0-android was computed. net9.0-browser was computed. net9.0-ios was computed. net9.0-maccatalyst was computed. net9.0-macos was computed. net9.0-tvos was computed. net9.0-windows was computed. net10.0 was computed. 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. |
-
net8.0
- Microsoft.CodeAnalysis.CSharp (>= 4.13.0)
- protobuf-net (>= 3.2.30)
- RandN (>= 0.4.0)
- ZstdSharp.Port (>= 0.8.0)
NuGet packages (3)
Showing the top 3 NuGet packages that depend on EricLouchez.Shorokoo.Core:
| Package | Downloads |
|---|---|
|
EricLouchez.Shorokoo
Define, train, and run neural networks in pure C#. Meta-package that brings the Shorokoo runtime (Shorokoo.Core), ready-made layers (Shorokoo.Modules), and the [Module] source generator (Shorokoo.CodeGen). Add exactly one backend: Shorokoo.LinuxCPU, Shorokoo.LinuxGPU, Shorokoo.WinCPU, or Shorokoo.WinGPU. |
|
|
EricLouchez.Shorokoo.OnnxRuntime
ONNX Runtime backend glue for Shorokoo (managed, platform-neutral). You normally do not install this directly; install Shorokoo.LinuxCPU, Shorokoo.LinuxGPU, Shorokoo.WinCPU, or Shorokoo.WinGPU instead, which bring the native ONNX Runtime for your platform. |
|
|
EricLouchez.Shorokoo.Modules
Baseline neural-network library for Shorokoo: ready-made layers, loss functions, and optimizers (SGD, SGD+momentum, AdamW, ...) built from Shorokoo modules. |
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 0.1.7-dev | 56 | 6/16/2026 |
| 0.1.6-dev | 49 | 6/16/2026 |
| 0.1.4-preview.1 | 82 | 6/12/2026 |