Method GetBestCategory

Namespace

LMKit.TextAnalysis

Assembly

LM-Kit.NET.dll

GetBestCategory(IList<string>, string, bool, CancellationToken)

Classifies the specified plain text into one of the predefined categories.

public int GetBestCategory(IList<string> categories, string content, bool normalize = true, CancellationToken cancellationToken = default)

Parameters

categories IList<string>

A list of predefined category names. Each entry should describe the corresponding category.

content string

The plain text to be classified.

normalize bool

Optional. Specifies whether the input parameters should be normalized. The default is true.

cancellationToken CancellationToken

Optional. A CancellationToken that can be used to cancel the operation.

Returns

int

The index of the best matching category from the categories list, or -1 if no suitable category is found.

Examples

using LMKit.Model;
using LMKit.TextAnalysis;
using System;
using System.Collections.Generic;

class Example
{
    static void Main()
    {
        LM model = LM.LoadFromModelID("lmkit-tasks:4b-preview");
        Categorization categorization = new Categorization(model);
        var categories = new List<string> { "finance", "travel", "entertainment" };
        string text = "Stock prices soared on Wall Street.";

        int bestCategoryIndex = categorization.GetBestCategory(categories, text);

        if (bestCategoryIndex != -1)
        {
            Console.WriteLine($"Category: {categories[bestCategoryIndex]}");
        }
        else
        {
            Console.WriteLine("No suitable category found.");
        }

        Console.WriteLine($"Confidence: {categorization.Confidence}");
    }
}

Exceptions

OperationCanceledException

Thrown if the operation is canceled via the cancellationToken.

GetBestCategory(IList<string>, IList<string>, string, bool, CancellationToken)

Classifies the specified plain text into one of the predefined categories.

public int GetBestCategory(IList<string> categories, IList<string> categoryDescriptions, string content, bool normalize = true, CancellationToken cancellationToken = default)

Parameters

categories IList<string>

A list of predefined categories.

categoryDescriptions IList<string>

Optional descriptions for each category.

content string

The plain text to be classified.

normalize bool

Specifies whether normalization should occur.

cancellationToken CancellationToken

A CancellationToken for canceling the operation.

Returns

int

The index of the best matching category, or -1 if none match.

GetBestCategory(IList<string>, Attachment, bool, CancellationToken)

Classifies the content provided as an attachment into one of the predefined categories.

public int GetBestCategory(IList<string> categories, Attachment content, bool normalize = true, CancellationToken cancellationToken = default)

Parameters

categories IList<string>

A list of predefined category names.

content Attachment

The attachment to be classified. Supports various file formats including images, PDF documents, HTML files, and Microsoft Office formats. See Attachment for the complete list of supported formats.

normalize bool

Specifies whether input parameters should be normalized. The default is true.

cancellationToken CancellationToken

A CancellationToken for canceling the operation.

Returns

int

The index of the best matching category, or -1 if none match.

Examples

using LMKit.Model;
using LMKit.TextAnalysis;
using LMKit.Data;
using System;
using System.Collections.Generic;

class Example
{
    static void Main()
    {
        LM model = LM.LoadFromModelID("lmkit-tasks:4b-preview");
        Categorization categorization = new Categorization(model);
        var categories = new List<string> { "invoice", "receipt", "contract", "letter" };

        // Classify a PDF document
        var attachment = new Attachment("path/to/document.pdf");
        int bestCategoryIndex = categorization.GetBestCategory(categories, attachment);

        if (bestCategoryIndex != -1)
        {
            Console.WriteLine($"Document type: {categories[bestCategoryIndex]}");
        }
    }
}

Exceptions

InvalidModelException

Thrown when the attachment contains visual content requiring vision input but the underlying language model does not support vision.

GetBestCategory(IList<string>, ImageBuffer, bool, CancellationToken)

Classifies the content provided in an image into one of the predefined categories. The underlying language model must support vision input.

public int GetBestCategory(IList<string> categories, ImageBuffer image, bool normalize = true, CancellationToken cancellationToken = default)

Parameters

categories IList<string>

A list of predefined category names.

image ImageBuffer

The image to classify.

normalize bool

Specifies whether input parameters should be normalized. The default is true.

cancellationToken CancellationToken

A CancellationToken used to cancel the operation.

Returns

int

The zero-based index of the best matching category, or -1 if none match.

Exceptions

InvalidModelException

Thrown if the underlying model does not support vision.

GetBestCategory(IList<string>, IList<string>, Attachment, bool, CancellationToken)

Classifies the content provided as an attachment into one of the predefined categories.

public int GetBestCategory(IList<string> categories, IList<string> categoryDescriptions, Attachment content, bool normalize = true, CancellationToken cancellationToken = default)

Parameters

categories IList<string>

A list of predefined category names.

categoryDescriptions IList<string>

Optional descriptions for each category.

content Attachment

The attachment to be classified. Supports various file formats including images, PDF documents, HTML files, and Microsoft Office formats.

normalize bool

Specifies whether input parameters should be normalized. The default is true.

cancellationToken CancellationToken

A CancellationToken for canceling the operation.

Returns

int

The index of the best matching category, or -1 if none match.

Exceptions

InvalidModelException

Thrown when the attachment contains visual content requiring vision input but the underlying language model does not support vision.