System.Numerics.Tensors 10.0.0-preview.4.25258.110
About
Provides methods for performing mathematical operations over tensors represented as spans. These methods are accelerated to use SIMD (Single instruction, multiple data) operations supported by the CPU where available.
Key Features
- Numerical operations on tensors represented as
ReadOnlySpan<float> - Element-wise arithmetic: Add, Subtract, Multiply, Divide, Exp, Log, Cosh, Tanh, etc.
- Tensor arithmetic: CosineSimilarity, Distance, Dot, Normalize, Softmax, Sigmoid, etc.
How to Use
using System.Numerics.Tensors;
var movies = new[] {
new { Title="The Lion King", Embedding= new [] { 0.10022575f, -0.23998135f } },
new { Title="Inception", Embedding= new [] { 0.10327095f, 0.2563685f } },
new { Title="Toy Story", Embedding= new [] { 0.095857024f, -0.201278f } },
new { Title="Pulp Function", Embedding= new [] { 0.106827796f, 0.21676421f } },
new { Title="Shrek", Embedding= new [] { 0.09568083f, -0.21177962f } }
};
var queryEmbedding = new[] { 0.12217915f, -0.034832448f };
var top3MoviesTensorPrimitives =
movies
.Select(movie =>
(
movie.Title,
Similarity: TensorPrimitives.CosineSimilarity(queryEmbedding, movie.Embedding)
))
.OrderByDescending(movies => movies.Similarity)
.Take(3);
foreach (var movie in top3MoviesTensorPrimitives)
{
Console.WriteLine(movie);
}
Main Types
The main types provided by this library are:
System.Numerics.Tensors.TensorPrimitives
Additional Documentation
Feedback & Contributing
System.Numerics.Tensors is released as open source under the MIT license. Bug reports and contributions are welcome at the GitHub repository.
Showing the top 20 packages that depend on System.Numerics.Tensors.
| Packages | Downloads |
|---|---|
|
Microsoft.SemanticKernel.Core
Semantic Kernel core orchestration, runtime and functions.
This package is automatically installed by 'Microsoft.SemanticKernel' package with other useful packages.
Install this package manually only if you are selecting individual Semantic Kernel components.
|
1 |
https://go.microsoft.com/fwlink/?LinkID=799421
.NET Framework 4.6.2
- Microsoft.Bcl.Numerics (>= 10.0.0-preview.4.25258.110)
- System.Memory (>= 4.6.3)
- System.Numerics.Vectors (>= 4.6.1)
.NET 8.0
- No dependencies.
.NET 9.0
- No dependencies.
.NET 10.0
- No dependencies.
.NET Standard 2.0
- Microsoft.Bcl.Numerics (>= 10.0.0-preview.4.25258.110)
- System.Memory (>= 4.6.3)
- System.Numerics.Vectors (>= 4.6.1)