Table of Contents

Class VectorizationStateItem

Namespace
FoundationaLLM.Vectorization.Services.VectorizationStates
Assembly
FoundationaLLM.Vectorization.Engine.dll

Properties that are serialized in the Parquet file associated with the vectorization state. Combines properties from the TextPartition and the TextEmbeddingVector artifacts.

public class VectorizationStateItem
Inheritance
VectorizationStateItem
Inherited Members
Extension Methods

Properties

PipelineName

The name of the pipeline the item relates to.

[ParquetRequired]
public string? PipelineName { get; set; }

Property Value

string

Position

The position of the item in the list of vectorization artifacts.

[ParquetRequired]
public int Position { get; set; }

Property Value

int

TextEmbeddingVector

The content of the TextEmbeddingVector artifact.

public List<float>? TextEmbeddingVector { get; set; }

Property Value

List<float>

TextEmbeddingVectorHash

The MD5 of the TextEmbeddingVector artifact content in string format.

public string? TextEmbeddingVectorHash { get; set; }

Property Value

string

TextEmbeddingVectorSize

The size of the TextEmbeddingVector artifact content (number of embedding dimensions).

public int TextEmbeddingVectorSize { get; set; }

Property Value

int

TextPartitionContent

The content of the TextPartition artifact.

[ParquetRequired]
public string? TextPartitionContent { get; set; }

Property Value

string

TextPartitionHash

The MD5 hash of the TextPartition artifact content.

[ParquetRequired]
public string? TextPartitionHash { get; set; }

Property Value

string

TextPartitionSize

The size of the TextPartition artifact content.

[ParquetRequired]
public int TextPartitionSize { get; set; }

Property Value

int