Class BootstrapResult<T>
Represents the results of bootstrap validation for a machine learning model, containing R² metrics for training, validation, and test datasets.
public class BootstrapResult<T>Type Parameters
TThe numeric type used for R² values, typically float or double.
- Inheritance
-
BootstrapResult<T>
- Inherited Members
Remarks
Bootstrap validation is a resampling technique used to evaluate machine learning models by repeatedly sampling from the available data with replacement. This class stores the R² (coefficient of determination) values for different data splits in the bootstrap process. R² measures the proportion of variance in the dependent variable that is predictable from the independent variables, with values ranging from 0 to 1 where higher values indicate better model fit. The class uses generic type parameter T to support different numeric types for the R² values, such as float, double, or decimal.
For Beginners: This class stores how well a model performs on different parts of your data.
When evaluating machine learning models:
- It's important to know how well they perform on different datasets
- Bootstrap validation creates multiple samples from your data to test model stability
- R² (R-squared) is a common metric that measures how well your model explains the variation in the data
This class stores three R² values:
- Training R²: How well the model fits the data it was trained on
- Validation R²: How well the model performs on data used for tuning
- Test R²: How well the model generalizes to completely new data
These values help you understand:
- If your model is underfitting (low R² on all datasets)
- If your model is overfitting (high training R², much lower test R²)
- How stable your model's performance is across different data splits
Constructors
BootstrapResult()
Initializes a new instance of the BootstrapResult class with all R² values set to zero.
public BootstrapResult()Remarks
This constructor creates a new BootstrapResult instance and initializes all R² values (TrainingR2, ValidationR2, and TestR2) to zero using the numeric operations appropriate for the generic type T. It uses the MathHelper.GetNumericOperations method to obtain the appropriate numeric operations for the type T, which allows the class to work with different numeric types such as float, double, or decimal. This initialization ensures that the R² values start at a well-defined state before being updated with actual results from the bootstrap validation process.
For Beginners: This constructor creates a new result object with all values initialized to zero.
When a new BootstrapResult is created:
- All three R² values are set to zero
- The constructor uses MathHelper to handle different numeric types
- This provides a clean starting point before actual results are calculated
This initialization is important because:
- It ensures consistent behavior regardless of how the object is created
- It prevents potential issues with uninitialized values
- It makes the code more robust across different numeric types
You typically won't need to call this constructor directly, as it will be used internally by the bootstrap validation process.
Properties
TestR2
Gets or sets the R² value for the test dataset.
public T TestR2 { get; set; }Property Value
- T
The R² value for the test dataset, initialized to zero.
Remarks
This property represents the coefficient of determination (R²) for the test dataset, which is the portion of data set aside and not used during model training or validation. The test R² provides the most realistic estimate of how well the model will perform on completely new, unseen data. It is the final evaluation metric used to assess the model's generalization ability. The test R² should be reasonably close to the validation R² if the validation process was effective. A significant drop from validation R² to test R² may indicate that the validation process was not representative of real-world data or that the model was indirectly overfitted through repeated hyperparameter tuning based on the validation set.
For Beginners: This value shows how well your model performs on completely new data.
The test R²:
- Measures how well your model generalizes to data it has never seen before
- Provides the most realistic estimate of real-world performance
- Is the final evaluation metric for your model
This value is important because:
- It represents how your model will likely perform in production
- It's the most honest assessment of your model's capabilities
- It helps determine if your model is ready for deployment
For example, a test R² of 0.80 means your model can explain 80% of the variation in new data, which would generally be considered good performance.
TrainingR2
Gets or sets the R² value for the training dataset.
public T TrainingR2 { get; set; }Property Value
- T
The R² value for the training dataset, initialized to zero.
Remarks
This property represents the coefficient of determination (R²) for the training dataset, which is the portion of data used to fit the model parameters. The R² value measures how well the model explains the variance in the dependent variable, with values ranging from 0 to 1. A value of 1 indicates that the model perfectly explains all the variance, while a value of 0 indicates that the model explains none of the variance. The training R² is typically higher than validation or test R² because the model is optimized specifically for the training data. A very high training R² compared to validation and test R² may indicate overfitting.
For Beginners: This value shows how well your model fits the data it was trained on.
The training R²:
- Measures how well your model explains the variation in the training data
- Ranges from 0 (poor fit) to 1 (perfect fit)
- Is calculated using the same data used to build the model
This value is important because:
- It shows whether your model can capture patterns in the data
- It serves as a baseline to compare with validation and test performance
- Very high values (close to 1) might indicate overfitting
For example, a training R² of 0.85 means your model explains 85% of the variation in the training data, which generally indicates a good fit.
ValidationR2
Gets or sets the R² value for the validation dataset.
public T ValidationR2 { get; set; }Property Value
- T
The R² value for the validation dataset, initialized to zero.
Remarks
This property represents the coefficient of determination (R²) for the validation dataset, which is the portion of data used to tune hyperparameters and evaluate the model during training. The validation R² provides an estimate of how well the model will perform on unseen data. It is typically lower than the training R² but should be reasonably close if the model is not overfitting. A significant drop from training R² to validation R² may indicate that the model is overfitting to the training data and not generalizing well. In bootstrap validation, the validation set is typically created by sampling with replacement from the original dataset.
For Beginners: This value shows how well your model performs on data used for tuning.
The validation R²:
- Measures how well your model generalizes to data not used in training
- Helps detect overfitting (when a model performs well on training data but poorly on new data)
- Is used to guide model selection and hyperparameter tuning
This value is important because:
- It provides feedback during the model development process
- It helps you choose between different model configurations
- It gives an early indication of how well the model might perform in production
For example, if your training R² is 0.85 but validation R² is only 0.65, this suggests your model might be overfitting to the training data.