Table of Contents

Class DiT<T>

Namespace
AiDotNet.Document.LayoutAware
Assembly
AiDotNet.dll

DiT (Document Image Transformer) for document image understanding.

public class DiT<T> : DocumentNeuralNetworkBase<T>, INeuralNetworkModel<T>, INeuralNetwork<T>, IInterpretableModel<T>, IInputGradientComputable<T>, IDisposable, ILayoutDetector<T>, IDocumentClassifier<T>, IDocumentModel<T>, IFullModel<T, Tensor<T>, Tensor<T>>, IModel<Tensor<T>, Tensor<T>, ModelMetadata<T>>, IModelSerializer, ICheckpointableModel, IParameterizable<T, Tensor<T>, Tensor<T>>, IFeatureAware, IFeatureImportance<T>, ICloneable<IFullModel<T, Tensor<T>, Tensor<T>>>, IGradientComputable<T, Tensor<T>, Tensor<T>>, IJitCompilable<T>

Type Parameters

T

The numeric type used for calculations.

Inheritance
DiT<T>
Implements
IFullModel<T, Tensor<T>, Tensor<T>>
IModel<Tensor<T>, Tensor<T>, ModelMetadata<T>>
IParameterizable<T, Tensor<T>, Tensor<T>>
ICloneable<IFullModel<T, Tensor<T>, Tensor<T>>>
IGradientComputable<T, Tensor<T>, Tensor<T>>
Inherited Members
Extension Methods

Remarks

DiT applies self-supervised pre-training on large-scale document images using a Vision Transformer (ViT) backbone, enabling strong document layout analysis without requiring OCR annotations.

For Beginners: DiT learns document understanding from images alone: 1. Pre-trains on 42 million document images 2. Uses masked image modeling (predicts missing patches) 3. Learns document-specific visual patterns

Key features:

  • Pure vision approach (no OCR needed for pre-training)
  • ViT-base/large architectures
  • State-of-the-art on document classification
  • Strong layout analysis performance

Example usage:

var model = new DiT<float>(architecture);
var result = model.DetectLayout(documentImage);

Reference: "DiT: Self-supervised Pre-training for Document Image Transformer" (ACM MM 2022) https://arxiv.org/abs/2203.02378

Constructors

DiT(NeuralNetworkArchitecture<T>, int, int, int, int, int, int, string, IOptimizer<T, Tensor<T>, Tensor<T>>?, ILossFunction<T>?)

Creates a DiT model using native layers for training and inference.

public DiT(NeuralNetworkArchitecture<T> architecture, int numClasses = 16, int imageSize = 224, int patchSize = 16, int hiddenDim = 768, int numLayers = 12, int numHeads = 12, string modelSize = "base", IOptimizer<T, Tensor<T>, Tensor<T>>? optimizer = null, ILossFunction<T>? lossFunction = null)

Parameters

architecture NeuralNetworkArchitecture<T>
numClasses int
imageSize int
patchSize int
hiddenDim int
numLayers int
numHeads int
modelSize string
optimizer IOptimizer<T, Tensor<T>, Tensor<T>>
lossFunction ILossFunction<T>

Remarks

Default Configuration (DiT-Base from ACM MM 2022): - Architecture: ViT-Base - Hidden dimension: 768 - Layers: 12, Heads: 12 - Patch size: 16×16 - Image size: 224×224 - Pre-training: Masked image modeling on IIT-CDIP

DiT(NeuralNetworkArchitecture<T>, string, int, int, int, int, int, int, string, IOptimizer<T, Tensor<T>, Tensor<T>>?, ILossFunction<T>?)

Creates a DiT model using a pre-trained ONNX model for inference.

public DiT(NeuralNetworkArchitecture<T> architecture, string onnxModelPath, int numClasses = 16, int imageSize = 224, int patchSize = 16, int hiddenDim = 768, int numLayers = 12, int numHeads = 12, string modelSize = "base", IOptimizer<T, Tensor<T>, Tensor<T>>? optimizer = null, ILossFunction<T>? lossFunction = null)

Parameters

architecture NeuralNetworkArchitecture<T>
onnxModelPath string
numClasses int
imageSize int
patchSize int
hiddenDim int
numLayers int
numHeads int
modelSize string
optimizer IOptimizer<T, Tensor<T>, Tensor<T>>
lossFunction ILossFunction<T>

Properties

AvailableCategories

Gets the available classification categories for this model.

public IReadOnlyList<string> AvailableCategories { get; }

Property Value

IReadOnlyList<string>

ExpectedImageSize

Gets the expected input image size (assumes square images).

public int ExpectedImageSize { get; }

Property Value

int

Remarks

Common values: 224 (ViT base), 384, 448, 512, 768, 1024. Input images will be resized to [ImageSize x ImageSize] before processing.

ModelSize

Gets the model size variant (base/large).

public string ModelSize { get; }

Property Value

string

PatchSize

Gets the patch size for the ViT backbone.

public int PatchSize { get; }

Property Value

int

RequiresOCR

Gets whether this model requires OCR preprocessing.

public override bool RequiresOCR { get; }

Property Value

bool

Remarks

Layout-aware models (LayoutLM, etc.) require OCR to provide text and bounding boxes. OCR-free models (Donut, Pix2Struct) process raw pixels directly.

SupportedDocumentTypes

Gets the supported document types for this model.

public override DocumentType SupportedDocumentTypes { get; }

Property Value

DocumentType

SupportedElementTypes

Gets the layout element types this detector can identify.

public IReadOnlyList<LayoutElementType> SupportedElementTypes { get; }

Property Value

IReadOnlyList<LayoutElementType>

Methods

ApplyDefaultPostprocessing(Tensor<T>)

Applies DiT's industry-standard postprocessing: pass-through (classification outputs are already final).

protected override Tensor<T> ApplyDefaultPostprocessing(Tensor<T> modelOutput)

Parameters

modelOutput Tensor<T>

Returns

Tensor<T>

ApplyDefaultPreprocessing(Tensor<T>)

Applies DiT's industry-standard preprocessing: ImageNet normalization.

protected override Tensor<T> ApplyDefaultPreprocessing(Tensor<T> rawImage)

Parameters

rawImage Tensor<T>

Returns

Tensor<T>

Remarks

DiT (Document Image Transformer) uses ImageNet normalization with mean=[0.485, 0.456, 0.406] and std=[0.229, 0.224, 0.225] (Microsoft paper).

ClassifyDocument(Tensor<T>)

Classifies a document image into predefined categories.

public DocumentClassificationResult<T> ClassifyDocument(Tensor<T> documentImage)

Parameters

documentImage Tensor<T>

The document image tensor.

Returns

DocumentClassificationResult<T>

Classification result with predicted category and confidence.

ClassifyDocument(Tensor<T>, int)

Classifies a document and returns top-K predictions.

public DocumentClassificationResult<T> ClassifyDocument(Tensor<T> documentImage, int topK)

Parameters

documentImage Tensor<T>

The document image tensor.

topK int

Number of top predictions to return.

Returns

DocumentClassificationResult<T>

Classification result with top-K predictions.

ClassifyDocumentBatch(IEnumerable<Tensor<T>>) 

public IEnumerable<DocumentClassificationResult<T>> ClassifyDocumentBatch(IEnumerable<Tensor<T>> documentImages)

Parameters

documentImages IEnumerable<Tensor<T>>

Returns

IEnumerable<DocumentClassificationResult<T>>

CreateNewInstance()

Creates a new instance of the same type as this neural network.

protected override IFullModel<T, Tensor<T>, Tensor<T>> CreateNewInstance()

Returns

IFullModel<T, Tensor<T>, Tensor<T>>

A new instance of the same neural network type.

Remarks

For Beginners: This creates a blank version of the same type of neural network.

It's used internally by methods like DeepCopy and Clone to create the right type of network before copying the data into it.

DeserializeNetworkSpecificData(BinaryReader)

Deserializes network-specific data that was not covered by the general deserialization process.

protected override void DeserializeNetworkSpecificData(BinaryReader reader)

Parameters

reader BinaryReader

The BinaryReader to read the data from.

Remarks

This method is called at the end of the general deserialization process to allow derived classes to read any additional data specific to their implementation.

For Beginners: Continuing the suitcase analogy, this is like unpacking that special compartment. After the main deserialization method has unpacked the common items (layers, parameters), this method allows each specific type of neural network to unpack its own unique items that were stored during serialization.

DetectLayout(Tensor<T>)

Detects layout regions in a document image.

public DocumentLayoutResult<T> DetectLayout(Tensor<T> documentImage)

Parameters

documentImage Tensor<T>

The document image tensor [batch, channels, height, width].

Returns

DocumentLayoutResult<T>

Layout detection result with regions and their types.

DetectLayout(Tensor<T>, double)

Detects layout regions with a specified confidence threshold.

public DocumentLayoutResult<T> DetectLayout(Tensor<T> documentImage, double confidenceThreshold)

Parameters

documentImage Tensor<T>

The document image tensor.

confidenceThreshold double

Minimum confidence for detected regions (0.0 to 1.0).

Returns

DocumentLayoutResult<T>

Filtered layout detection result.

Remarks

Higher thresholds return fewer but more confident detections. Lower thresholds return more detections but may include false positives.

Dispose(bool)

Disposes of resources used by this model.

protected override void Dispose(bool disposing)

Parameters

disposing bool

True if disposing managed resources.

EncodeDocument(Tensor<T>)

Processes a document image and returns encoded features.

public Tensor<T> EncodeDocument(Tensor<T> documentImage)

Parameters

documentImage Tensor<T>

The document image tensor [batch, channels, height, width] or [channels, height, width].

Returns

Tensor<T>

Encoded document features suitable for downstream tasks.

Remarks

For Beginners: This method converts a document image into a numerical representation (feature vector) that captures the document's content and structure. These features can then be used for tasks like classification, QA, or information extraction.

GetModelMetadata()

Gets the metadata for this neural network model.

public override ModelMetadata<T> GetModelMetadata()

Returns

ModelMetadata<T>

A ModelMetaData object containing information about the model.

GetModelSummary()

Gets a summary of the model architecture.

public string GetModelSummary()

Returns

string

A string describing the model's architecture, parameters, and capabilities.

InitializeLayers()

Initializes the layers of the neural network based on the architecture.

protected override void InitializeLayers()

Remarks

For Beginners: This method sets up all the layers in your neural network according to the architecture you've defined. It's like assembling the parts of your network before you can use it.

Predict(Tensor<T>)

Makes a prediction using the neural network.

public override Tensor<T> Predict(Tensor<T> input)

Parameters

input Tensor<T>

The input data to process.

Returns

Tensor<T>

The network's prediction.

Remarks

For Beginners: This is the main method you'll use to get results from your trained neural network. You provide some input data (like an image or text), and the network processes it through all its layers to produce an output (like a classification or prediction).

SerializeNetworkSpecificData(BinaryWriter)

Serializes network-specific data that is not covered by the general serialization process.

protected override void SerializeNetworkSpecificData(BinaryWriter writer)

Parameters

writer BinaryWriter

The BinaryWriter to write the data to.

Remarks

This method is called at the end of the general serialization process to allow derived classes to write any additional data specific to their implementation.

For Beginners: Think of this as packing a special compartment in your suitcase. While the main serialization method packs the common items (layers, parameters), this method allows each specific type of neural network to pack its own unique items that other networks might not have.

Train(Tensor<T>, Tensor<T>)

Trains the neural network on a single input-output pair.

public override void Train(Tensor<T> input, Tensor<T> expectedOutput)

Parameters

input Tensor<T>

The input data.

expectedOutput Tensor<T>

The expected output for the given input.

Remarks

This method performs one training step on the neural network using the provided input and expected output. It updates the network's parameters to reduce the error between the network's prediction and the expected output.

For Beginners: This is how your neural network learns. You provide: - An input (what the network should process) - The expected output (what the correct answer should be)

The network then:

  1. Makes a prediction based on the input
  2. Compares its prediction to the expected output
  3. Calculates how wrong it was (the loss)
  4. Adjusts its internal values to do better next time

After training, you can get the loss value using the GetLastLoss() method to see how well the network is learning.

UpdateParameters(Vector<T>)

Updates the network's parameters with new values.

public override void UpdateParameters(Vector<T> gradients)

Parameters

gradients Vector<T>

Remarks

For Beginners: During training, a neural network's internal values (parameters) get adjusted to improve its performance. This method allows you to update all those values at once by providing a complete set of new parameters.

This is typically used by optimization algorithms that calculate better parameter values based on training data.

ValidateInputShape(Tensor<T>)

Validates that an input tensor has the correct shape for this model.

public void ValidateInputShape(Tensor<T> documentImage)

Parameters

documentImage Tensor<T>

The tensor to validate.

Exceptions

ArgumentException

Thrown if the tensor shape is invalid.

Edit this page