Class DataSetStats<T, TInput, TOutput>

Namespace

AiDotNet.Models

Assembly

AiDotNet.dll

public class DataSetStats<T, TInput, TOutput>

Type Parameters

T

TInput

TOutput

Inheritance

object

DataSetStats<T, TInput, TOutput>

Inherited Members

object.Equals(object)

object.Equals(object, object)

object.GetHashCode()

object.GetType()

object.MemberwiseClone()

object.ReferenceEquals(object, object)

object.ToString()

Constructors

DataSetStats()

Initializes a new instance of the DataSetStats class with default empty model data.

public DataSetStats()

Properties

Actual

Gets or sets the actual target values.

public TOutput Actual { get; set; }

Property Value

TOutput

The actual target values of type TOutput.

Remarks

This property contains the actual target values for the dataset. The structure of this object matches that of the Predicted property and depends on the type of model and prediction task. These values represent the true outcomes that the model attempts to predict.

[Existing remarks for Actual]

ActualBasicStats

Gets or sets the basic descriptive statistics for the actual target values.

public BasicStats<T> ActualBasicStats { get; set; }

Property Value

BasicStats<T>

A BasicStats<T> object containing descriptive statistics for the actual values.

Remarks

[Existing remarks for ActualBasicStats]

ErrorStats

Gets or sets the error statistics for the model's predictions.

public ErrorStats<T> ErrorStats { get; set; }

Property Value

ErrorStats<T>

An ErrorStats<T> object containing various error metrics.

Remarks

[Existing remarks for ErrorStats]

Features

Gets or sets the input features.

public TInput Features { get; set; }

Property Value

TInput

The input features of type TInput.

Remarks

This property contains the input feature data for the dataset. The structure of this object depends on the type of model and input data. For traditional machine learning tasks, it might be a matrix where each row represents an observation and each column a feature. For more complex tasks like image or sequence processing, it might be a tensor or other multi-dimensional structure.

[Existing remarks for Features]

IsDataProvided

Gets or sets a value indicating whether data was actually provided for this dataset.

public bool IsDataProvided { get; set; }

Property Value

bool

True if data was provided, false if this is a placeholder for missing data.

Remarks

This property distinguishes between "data was never provided" (false) and "data was provided but may be empty" (true). This is important for evaluation because an empty DataSetStats could mean either scenario, and users need to know whether metrics are unavailable due to missing data or due to actual empty data.

For Beginners: This flag tells you whether this dataset was actually evaluated. If false, it means the data was never provided (e.g., no validation set was given), so the statistics in this object are not meaningful.

Predicted

Gets or sets the predicted values.

public TOutput Predicted { get; set; }

Property Value

TOutput

The predicted values of type TOutput.

Remarks

This property contains the model's predictions for the dataset. The structure of this object depends on the type of model and prediction task. For regression tasks, it might be a vector where each element corresponds to an observation in the dataset. For more complex tasks like image classification or sequence prediction, it might be a tensor or other multi-dimensional structure.

[Existing remarks for Predicted]

PredictedBasicStats

Gets or sets the basic descriptive statistics for the predicted values.

public BasicStats<T> PredictedBasicStats { get; set; }

Property Value

BasicStats<T>

A BasicStats<T> object containing descriptive statistics for the predicted values.

Remarks

[Existing remarks for PredictedBasicStats]

PredictionStats

Gets or sets the prediction quality statistics for the model.

public PredictionStats<T> PredictionStats { get; set; }

Property Value

PredictionStats<T>

A PredictionStats<T> object containing various prediction quality metrics.

Remarks

[Existing remarks for PredictionStats]

RobustnessStats

Gets or sets adversarial robustness diagnostics for the dataset.

public RobustnessStats<T> RobustnessStats { get; set; }

Property Value

RobustnessStats<T>

Remarks

This is populated when adversarial robustness evaluation is enabled. It contains metrics such as clean accuracy, adversarial accuracy, certified accuracy, and attack success rates.

For Beginners: This stores information about how well the model resists adversarial attacks. Adversarial attacks are specially crafted inputs designed to fool machine learning models. High adversarial accuracy means the model is robust against such attacks.

UncertaintyStats

Gets or sets uncertainty quantification diagnostics for the dataset.

public UncertaintyStats<T> UncertaintyStats { get; set; }

Property Value

UncertaintyStats<T>

Remarks

This is populated when uncertainty quantification is enabled and the evaluation flow requests UQ diagnostics.