SmartExtensions

Namespace: Wisej.AI

Assembly: Wisej.AI

Provides extension methods for various operations such as calculating cosine similarity, computing clusters, converting strings to camel case, and managing services in a service provider.

C#

public class SmartExtensions

VB.NET

Public Class SmartExtensions

Methods

AddOrReplaceService<TService, TImplementation>(services, lifetime)

Adds or replaces a service in the service provider with the specified implementation type.

Parameter	Type	Description
TService		The type of service to add or replace.
TImplementation		The type of the implementation to use.
services	ServiceProvider	The service provider to modify.
lifetime	ServiceLifetime	The lifetime of the service.

Returns: ServiceProvider. The modified service provider.

AddOrReplaceService<TService>(services, implementation, lifetime)

Adds or replaces a service in the service provider with the specified implementation instance.

Parameter	Type	Description
TService		The type of service to add or replace.
services	ServiceProvider	The service provider to modify.
implementation	TService	The implementation instance to use.
lifetime	ServiceLifetime	The lifetime of the service.

Returns: ServiceProvider. The modified service provider.

AddOrReplaceService<TService>(services, factory, lifetime)

Adds or replaces a service in the service provider using a factory method.

Parameter	Type	Description
TService		The type of service to add or replace.
services	ServiceProvider	The service provider to modify.
factory	Func<Type, Object>	The factory method to create the service instance.
lifetime	ServiceLifetime	The lifetime of the service.

Returns: ServiceProvider. The modified service provider.

ComputeClusters(embeddings, count, maxDivergence)

Computes clusters from the given embeddings using the K-Means clustering algorithm.

Parameter	Type	Description
embeddings	Single[][]	The embeddings to cluster.
count	Int32	The number of clusters to create.
maxDivergence	Double	The maximum divergence allowed for convergence.

Returns: ValueTuple`2[]. An array of tuples containing the centroid and vectors of each cluster.

K-Means clustering is a method of vector quantization, originally from signal processing, that is popular for cluster analysis in data mining. This method partitions the embeddings into count clusters.

CosineSimilarity(vectorA, vectorB)

Calculates the cosine similarity between two vectors.

Parameter	Type	Description
vectorA	Single[]	The first vector.
vectorB	Single[]	The second vector.

Returns: Single. The cosine similarity between the two vectors.

Cosine similarity is a measure of similarity between two non-zero vectors of an inner product space. It is defined as the cosine of the angle between the two vectors.

CosineSimilarity(vectorA, vectorB)

Calculates the cosine similarity between a vector and an array of vectors.

Parameter	Type	Description
vectorA	Single[]	The vector to compare.
vectorB	Single[][]	The array of vectors to compare against.

Returns: Single[]. An array of cosine similarity values for each vector in the array.

This method extends the single vector cosine similarity calculation to handle multiple vectors, returning an array of similarity scores.

ToCamelCase(text)

Converts the first character of the string to lowercase, making it camel case.

Parameter	Type	Description
text	String	The string to convert.

Returns: String. The camel case version of the string.

This method is useful for converting PascalCase strings to camelCase, which is often used in JSON serialization and other contexts.