Svr / Regressor Layer

Support Vector Regression (SVR): Advanced regression with kernels.

Mathematical formulation: $f (x) = i = 1 \sum n (α_{i} - α_{i}^{*}) K (x_{i}, x) + b$ where:

αᵢ, αᵢ* are Lagrange multipliers
K(x,y) is the kernel function
b is the bias term

Key characteristics:

ε-insensitive loss function
Kernel-based learning
Robust to outliers
Non-linear modeling

Advantages:

High prediction accuracy
Good generalization
Handles non-linearity
Robust performance

Common applications:

Financial forecasting
Function approximation
System modeling
Time series prediction

Outputs:

Predicted Table: Results with predictions
Validation Results: Cross-validation metrics
Test Metric: Hold-out performance
Feature Importances: Support vector weights

Note: Computational complexity is O(n²). For large datasets (>10,000 samples), consider LinearSVR or SGDRegressor instead.

Table

Predicted Table

Validation Results

Test Metric

Feature Importances

SelectFeatures

[column, ...]

Feature column selection for SVR:

Requirements:

Data properties:
- Numeric values only
- No missing values
- Finite numbers
- Comparable scales
Preprocessing needs:
- Standardization/scaling crucial
- Outlier handling
- Feature correlation check
- Missing value treatment
SVR considerations:
- Feature relevance
- Kernel compatibility
- Dimensionality impact
- Computational cost
Best practices:
- Scale to [-1, 1] or [0, 1]
- Remove redundant features
- Handle outliers
- Check correlations

Note: If empty, uses all numeric columns except target

SelectTarget

column

Target column specification for SVR:

Requirements:

Data type:
- Numeric continuous
- No missing values
- Finite numbers
- Real-valued
Statistical properties:
- Scale consideration
- Distribution check
- Outlier presence
- Noise level
SVR specifics:
- ε-tube compatibility
- Error tolerance
- Prediction range
- Loss function scale
Preprocessing:
- Scaling recommended
- Outlier treatment
- Transform if needed
- Range normalization

Note: Must be a single numeric column

Params

oneof

DefaultParams

Default SVR configuration:

Model structure:
- C = 1.0 (regularization strength)
- RBF kernel (versatile default)
- ε = 0.1 (error tube width)
Kernel settings:
- γ = 'scale' (adaptive radius)
- degree = 3 (polynomial)
- coef₀ = 0.0 (independent term)
Optimization parameters:
- Cache = 200MB (memory usage)
- tol = 0.001 (convergence)
- Shrinking = true (speed)
- max_iter = -1 (unlimited)

Best suited for:

Initial modeling
Medium-sized datasets
Unknown patterns
General regression tasks

CustomParams

Customizable SVR parameters:

Parameter categories:

Model complexity:
- Regularization (C)
- Kernel selection
- Error tolerance (ε)
Kernel configuration:
- Function type
- Shape parameters
- Feature mapping
Optimization control:
- Memory usage
- Convergence criteria
- Algorithm behavior

Trade-offs:

Accuracy vs complexity
Speed vs precision
Memory vs computation

CFactor

f64

Regularization parameter (C):

Impact on model:

Large C: High variance, low bias
Small C: Low variance, high bias

Selection guide:

Small: 0.1-1.0 (more regularization)
Medium: 1.0-10.0 (balanced)
Large: 10.0-100.0 (less regularization)

Considerations:

Noise level
Training size
Outlier presence
Model complexity

Kernel

enum

Rbf

Kernel functions for SVR:

Linear: K(x,y) = x·y
- Fastest computation
- Linear relationships
- High-dimensional data
Polynomial: K(x,y) = (γx·y + coef₀)^degree
- Feature interactions
- Degree controls complexity
- Useful for normalized data
RBF: K(x,y) = exp(-γ||x-y||²)
- Most versatile kernel
- Infinite dimensions
- Local influence
Sigmoid: K(x,y) = tanh(γx·y + coef₀)
- Neural network relation
- S-shaped responses
- Binary patterns

Selection impact:

Model complexity
Training time
Prediction accuracy
Generalization ability

Linear ~

Linear kernel function:

Formula: $K (x, y) = x^{T} y$

Properties:

Simplest kernel
Fast computation
Memory efficient
Linear separation

Best for:

High-dimensional data
Text classification
Sparse features
Linear relationships

Poly ~

Polynomial kernel function:

Formula: $K (x, y) = (γ x^{T} y + coe f_{0})^{d e g ree}$

Properties:

Feature interactions
Controlled complexity
Bounded response
Global influence

Best for:

Feature combinations
Normalized data
Moderate non-linearity
Pattern recognition

Rbf ~

Radial Basis Function kernel:

Formula: $K (x, y) = e x p (- γ ∣∣ x - y ∣ ∣^{2})$

Properties:

Infinite dimensions
Local sensitivity
Universal approximator
Distance-based

Best for:

Unknown relationships
Non-linear patterns
Continuous features
General-purpose use

Sigmoid ~

Sigmoid kernel function:

Formula: $K (x, y) = t anh (γ x^{T} y + coe f_{0})$

Properties:

Neural network relation
S-shaped response
Non-monotonic
Bounded output

Best for:

Neural network alternative
Binary patterns
Signal processing
Specific non-linearities

Degree

u32

Polynomial kernel degree:

Effect on learning:

Controls feature interactions
Affects model complexity
Impacts training time

Common values:

2: Quadratic relationships
3: Cubic patterns (default)
4+: Higher-order interactions

Note: Only used with polynomial kernel

Gamma

enum

Scale

Kernel coefficient strategies:

Mathematical impact: $K (x, y) = f (γ ∣∣ x - y ∣ ∣^{2})$

Role in kernels:

RBF: Controls radius of influence
Polynomial: Scales inner product
Sigmoid: Affects slope

Selection guidelines:

Small γ: Large influence radius
Large γ: Small influence radius
Optimal γ: Data-dependent

Impact on learning:

Feature importance
Model complexity
Training stability
Generalization

Scale ~

Scale-dependent gamma:

Formula: $γ = \frac{1}{n _{f e a t u res} \cdot Va r ( X )}$

Properties:

Data-adaptive scaling
Variance-aware
Feature-normalized
Robust behavior

Advantages:

Handles different scales
Modern default choice
Automatic adaptation
Stable performance

Auto ~

Feature-based gamma:

Formula: $γ = \frac{1}{n _{f e a t u res}}$

Properties:

Dimension-based scaling
Simple heuristic
Scale-independent
Legacy default

Best for:

Normalized features
Similar scales
Historical compatibility
Quick baselines

Custom ~

User-defined gamma:

Configuration:

Manual value setting
Expert knowledge input
Problem-specific tuning
Fine control

Use cases:

Known domain requirements
Cross-validation results
Performance optimization
Research experiments

GammaF

f64

Custom gamma value:

Impact on kernels:

RBF: Influence radius
Poly: Feature scaling
Sigmoid: Response slope

Typical ranges:

Small: 0.0001-0.001 (wide influence)
Medium: 0.001-0.1 (balanced)
Large: 0.1-1.0 (narrow influence)

Note: Only used when Gamma='Custom'

Coef0

f64

Independent term in kernel:

Usage in kernels:

Polynomial: Offset term
Sigmoid: Threshold

Impact:

Controls model flexibility
Affects feature interactions
Influences decision boundary

Common range: [-1.0, 1.0]

Shrinking

bool

true

Whether to use the shrinking heuristic.

When enabled:

Removes bounded variables
Speeds up optimization
Reduces memory usage

Trade-offs:

Speed vs precision
Memory vs accuracy
Training efficiency

Epsilon

f64

0.1

ε-tube width parameter:

Definition:

Defines insensitive region
Controls prediction precision
Affects support vector count

Selection guide:

Small ε: High precision, more SVs
Large ε: Lower precision, fewer SVs

Typical range: [0.01, 0.5]

Tol

f64

0.001

Optimization tolerance:

Controls:

Convergence precision
Training duration
Solution accuracy

Common values:

Strict: 1e-4 or smaller
Standard: 1e-3 (default)
Relaxed: 1e-2 or larger

CacheSize

u64

200

Kernel cache size (MB):

Impact:

Training speed
Memory usage
Computation efficiency

Guidelines:

Small: 50-100MB
Medium: 200MB (default)
Large: 500MB+

Trade-off: Speed vs memory

MaxIter

i64

-1

Maximum iterations limit:

Settings:

-1: No limit (default)
>0: Maximum iterations

Purpose:

Controls training time
Prevents endless loops
Resource management

Note: May affect convergence

GridSearch

SVR hyperparameter optimization:

Search space organization:

Model complexity:
- Regularization ranges
- Kernel selection
- Error tolerance
Kernel configuration:
- Function types
- Shape parameters
- Scale factors
Optimization settings:
- Convergence criteria
- Resource limits
- Algorithm behavior

Best practices:

Start with coarse grid
Refine promising regions
Consider computation cost
Monitor resource usage

CFactor

[f64, ...]

Regularization parameter search:

Common search patterns:

Logarithmic scale:
- [0.1, 1.0, 10.0, 100.0]
- Wide exploration
Fine-tuning:
- [0.8, 1.0, 1.2]
- Around promising value
Problem-specific:
- Based on data scale
- Noise sensitivity
- Model complexity needs

Kernel

[enum, ...]

Rbf

Kernel functions for SVR:

Linear: K(x,y) = x·y
- Fastest computation
- Linear relationships
- High-dimensional data
Polynomial: K(x,y) = (γx·y + coef₀)^degree
- Feature interactions
- Degree controls complexity
- Useful for normalized data
RBF: K(x,y) = exp(-γ||x-y||²)
- Most versatile kernel
- Infinite dimensions
- Local influence
Sigmoid: K(x,y) = tanh(γx·y + coef₀)
- Neural network relation
- S-shaped responses
- Binary patterns

Selection impact:

Model complexity
Training time
Prediction accuracy
Generalization ability

Linear ~

Linear kernel function:

Formula: $K (x, y) = x^{T} y$

Properties:

Simplest kernel
Fast computation
Memory efficient
Linear separation

Best for:

High-dimensional data
Text classification
Sparse features
Linear relationships

Poly ~

Polynomial kernel function:

Formula: $K (x, y) = (γ x^{T} y + coe f_{0})^{d e g ree}$

Properties:

Feature interactions
Controlled complexity
Bounded response
Global influence

Best for:

Feature combinations
Normalized data
Moderate non-linearity
Pattern recognition

Rbf ~

Radial Basis Function kernel:

Formula: $K (x, y) = e x p (- γ ∣∣ x - y ∣ ∣^{2})$

Properties:

Infinite dimensions
Local sensitivity
Universal approximator
Distance-based

Best for:

Unknown relationships
Non-linear patterns
Continuous features
General-purpose use

Sigmoid ~

Sigmoid kernel function:

Formula: $K (x, y) = t anh (γ x^{T} y + coe f_{0})$

Properties:

Neural network relation
S-shaped response
Non-monotonic
Bounded output

Best for:

Neural network alternative
Binary patterns
Signal processing
Specific non-linearities

Degree

[u32, ...]

Polynomial degree search:

Search spaces:

Standard range:
- [2, 3, 4]
- Common patterns
Extended:
- [2, 3, 4, 5]
- Higher complexity
Specific:
- Based on domain
- Known relationships

Note: For polynomial kernel

Gamma

[enum, ...]

Scale

Kernel coefficient strategies:

Mathematical impact: $K (x, y) = f (γ ∣∣ x - y ∣ ∣^{2})$

Role in kernels:

RBF: Controls radius of influence
Polynomial: Scales inner product
Sigmoid: Affects slope

Selection guidelines:

Small γ: Large influence radius
Large γ: Small influence radius
Optimal γ: Data-dependent

Impact on learning:

Feature importance
Model complexity
Training stability
Generalization

Scale ~

Scale-dependent gamma:

Formula: $γ = \frac{1}{n _{f e a t u res} \cdot Va r ( X )}$

Properties:

Data-adaptive scaling
Variance-aware
Feature-normalized
Robust behavior

Advantages:

Handles different scales
Modern default choice
Automatic adaptation
Stable performance

Auto ~

Feature-based gamma:

Formula: $γ = \frac{1}{n _{f e a t u res}}$

Properties:

Dimension-based scaling
Simple heuristic
Scale-independent
Legacy default

Best for:

Normalized features
Similar scales
Historical compatibility
Quick baselines

Custom ~

User-defined gamma:

Configuration:

Manual value setting
Expert knowledge input
Problem-specific tuning
Fine control

Use cases:

Known domain requirements
Cross-validation results
Performance optimization
Research experiments

GammaF

[f64, ...]

Custom gamma value search:

Search patterns:

Log scale:
- [0.0001, 0.001, 0.01, 0.1]
- Wide exploration
Fine-grained:
- Around best gamma
- Narrow range
Data-driven:
- Based on feature scales
- Problem characteristics

Coef0

[f64, ...]

Independent term search:

Search spaces:

Standard:
- [0.0, 0.5, 1.0]
- Basic range
Extended:
- [-1.0, 0.0, 1.0]
- Full range
Fine-tuning:
- Around best value
- Small increments

For poly and sigmoid kernels

Shrinking

[bool, ...]

true

Shrinking heuristic search:

Options:

Default: [true]
- Enable optimization
- Faster training
Compare: [true, false]
- Performance impact
- Speed vs precision
Specific:
- Based on dataset
- Resource constraints

Epsilon

[f64, ...]

0.1

ε-tube width search:

Search ranges:

Standard:
- [0.05, 0.1, 0.2]
- Common values
Precision focus:
- [0.01, 0.05, 0.1]
- Higher accuracy
Wide range:
- [0.01, 0.1, 0.5]
- Explore tolerance

Tol

[f64, ...]

0.001

Convergence tolerance search:

Search patterns:

Standard:
- [1e-4, 1e-3, 1e-2]
- Common range
High precision:
- [1e-5, 1e-4, 1e-3]
- Exact solutions
Quick convergence:
- [1e-3, 1e-2]
- Faster training

CacheSize

f64

200

Kernel cache memory size:

Guidelines:

Small: 50-100MB
- Limited memory
- Smaller datasets
Medium: 200MB
- Balanced choice
- Default setting
Large: 500MB+
- Fast training
- Large datasets

MaxIter

[i64, ...]

-1

Maximum iterations search:

Search options:

Unlimited: [-1]
- Full convergence
- No time limit
Limited: [1000, 2000, 5000]
- Time constrained
- Resource managed
Quick runs: [500, 1000]
- Fast iterations
- Initial testing

RefitScore

enum

R2score

Regression model evaluation metrics:

Purpose:

Model performance evaluation
Error measurement
Quality assessment
Model comparison

Selection criteria:

Error distribution
Scale sensitivity
Domain requirements
Business objectives

Default ~

Model's native scoring method:

Typically R² score
Model-specific implementation
Standard evaluation
Quick assessment

R2score ~

Coefficient of determination (R²):

Formula: $R^{2} = 1 - \frac{\sum ( y _{i} - y ^ _{i} ) ^{2}}{\sum ( y _{i} - y ˉ ) ^{2}}$

Properties:

Range: (-∞, 1]
1: Perfect prediction
0: Constant model
Negative: Worse than mean

Best for:

General performance
Variance explanation
Model comparison
Standard reporting

ExplainedVariance ~

Explained variance score:

Formula: $1 - \frac{Va r ( y - y ^ )}{Va r ( y )}$

Properties:

Range: (-∞, 1]
Accounts for bias
Variance focus
Similar to R²

Best for:

Variance analysis
Bias assessment
Model stability

MaxError ~

Maximum absolute error:

Formula: $max (∣ y_{i} - \overset{y}{^}_{i} ∣)$

Properties:

Worst case error
Original scale
Sensitive to outliers
Upper error bound

Best for:

Critical applications
Error bounds
Safety margins
Risk assessment

NegMeanAbsoluteError ~

Negative mean absolute error:

Formula: $- \frac{1}{n} \sum ∣ y_{i} - \overset{y}{^}_{i} ∣$

Properties:

Linear error scale
Robust to outliers
Original units
Negated for optimization

Best for:

Robust evaluation
Interpretable errors
Outlier presence

NegMeanSquaredError ~

Negative mean squared error:

Formula: $- \frac{1}{n} \sum (y_{i} - \overset{y}{^}_{i})^{2}$

Properties:

Squared error scale
Outlier sensitive
Squared units
Negated for optimization

Best for:

Standard optimization
Large error penalty
Statistical analysis

NegRootMeanSquaredError ~

Negative root mean squared error:

Formula: $- \frac{1}{n} \sum (y_{i} - \overset{y}{^}_{i})^{2}$

Properties:

Original scale
Outlier sensitive
Interpretable units
Negated for optimization

Best for:

Standard reporting
Interpretable errors
Model comparison

NegMeanSquaredLogError ~

Negative mean squared logarithmic error:

Formula: $- \frac{1}{n} \sum (lo g (1 + y_{i}) - lo g (1 + \overset{y}{^}_{i}))^{2}$

Properties:

Relative error scale
For positive values
Sensitive to ratios
Negated for optimization

Best for:

Exponential growth
Relative differences
Positive predictions

NegMedianAbsoluteError ~

Negative median absolute error:

Formula: $- m e d ian (∣ y_{i} - \overset{y}{^}_{i} ∣)$

Properties:

Highly robust
Original scale
Outlier resistant
Negated for optimization

Best for:

Robust evaluation
Heavy-tailed errors
Outlier presence

NegMeanPoissonDeviance ~

Negative Poisson deviance:

Formula: $- 2 \sum (y_{i} lo g (\frac{y _{i}}{y ^ _{i}}) - (y_{i} - \overset{y}{^}_{i}))$

Properties:

For count data
Non-negative values
Poisson assumption
Negated for optimization

Best for:

Count prediction
Event frequency
Rate modeling

NegMeanGammaDeviance ~

Negative Gamma deviance:

Formula: $- 2 \sum (lo g (\frac{y _{i}}{y ^ _{i}}) + \frac{y _{i} - y ^ _{i}}{y ^ _{i}})$

Properties:

For positive continuous data
Constant CV assumption
Relative errors
Negated for optimization

Best for:

Positive continuous data
Multiplicative errors
Financial modeling

NegMeanAbsolutePercentageError ~

Negative mean absolute percentage error:

Formula: $- \frac{100}{n} \sum ∣ \frac{y _{i} - y ^ _{i}}{y _{i}} ∣$

Properties:

Percentage scale
Scale independent
For non-zero targets
Negated for optimization

Best for:

Relative performance
Scale-free comparison
Business metrics

D2AbsoluteErrorScore ~

D² score with absolute error:

Formula: $1 - \frac{\sum ∣ y _{i} - y ^ _{i} ∣}{\sum ∣ y _{i} - y ˉ ∣}$

Properties:

Range: (-∞, 1]
Robust version of R²
Linear error scale
Outlier resistant

Best for:

Robust evaluation
Non-normal errors
Alternative to R²

D2PinballScore ~

D² score with pinball loss:

Properties:

Quantile focus
Asymmetric errors
Risk assessment
Distribution modeling

Best for:

Quantile regression
Risk analysis
Asymmetric costs
Distribution tails

D2TweedieScore ~

D² score with Tweedie deviance:

Properties:

Compound Poisson-Gamma
Flexible dispersion
Mixed distributions
Insurance modeling

Best for:

Insurance claims
Mixed continuous-discrete
Compound distributions
Specialized modeling

Split

oneof

DefaultSplit

Standard train-test split configuration optimized for general classification tasks.

Configuration:

Test size: 20% (0.2)
Random seed: 98
Shuffling: Enabled
Stratification: Based on target distribution

Advantages:

Preserves class distribution
Provides reliable validation
Suitable for most datasets

Best for:

Medium to large datasets
Independent observations
Initial model evaluation

Splitting uses the ShuffleSplit strategy or StratifiedShuffleSplit strategy depending on the field stratified. Note: If shuffle is false then stratified must be false.

CustomSplit

Configurable train-test split parameters for specialized requirements. Allows fine-tuning of data division strategy for specific use cases or constraints.

Use cases:

Time series data
Grouped observations
Specific train/test ratios
Custom validation schemes

RandomState

u64

Random seed for reproducible splits. Ensures:

Consistent train/test sets
Reproducible experiments
Comparable model evaluations

Same seed guarantees identical splits across runs.

Shuffle

bool

true

Data shuffling before splitting. Effects:

true: Randomizes order, better for i.i.d. data
false: Maintains order, important for time series

When to disable:

Time dependent data
Sequential patterns
Grouped observations

TrainSize

f64

0.8

Proportion of data for training. Considerations:

Larger (e.g., 0.8-0.9): Better model learning
Smaller (e.g., 0.5-0.7): Better validation

Common splits:

0.8: Standard (80/20 split)
0.7: More validation emphasis
0.9: More training emphasis

Stratified

bool

false

Maintain class distribution in splits. Important when:

Classes are imbalanced
Small classes present
Representative splits needed

Requirements:

Classification tasks only
Cannot use with shuffle=false
Sufficient samples per class

Cv

oneof

DefaultCv

Standard cross-validation configuration using stratified 3-fold splitting.

Configuration:

Folds: 3
Method: StratifiedKFold
Stratification: Preserves class proportions

Advantages:

Balanced evaluation
Reasonable computation time
Good for medium-sized datasets

Limitations:

May be insufficient for small datasets
Higher variance than larger fold counts
May miss some data patterns

CustomCv

Configurable stratified k-fold cross-validation for specific validation requirements.

Features:

Adjustable fold count with NFolds determining the number of splits.
Stratified sampling
Preserved class distributions

Use cases:

Small datasets (more folds)
Large datasets (fewer folds)
Detailed model evaluation
Robust performance estimation

NFolds

u32

Number of cross-validation folds. Guidelines:

3-5: Large datasets, faster training
5-10: Standard choice, good balance
10+: Small datasets, thorough evaluation

Trade-offs:

More folds: Better evaluation, slower training
Fewer folds: Faster training, higher variance

Must be at least 2.

KfoldCv

K-fold cross-validation without stratification. Divides data into k consecutive folds for iterative validation.

Process:

Splits data into k equal parts
Each fold serves as validation once
Remaining k-1 folds form training set

Use cases:

Regression problems
Large, balanced datasets
When stratification unnecessary
Continuous target variables

Limitations:

May not preserve class distributions
Less suitable for imbalanced data
Can create biased splits with ordered data

NSplits

u32

Number of folds for cross-validation. Selection guide: Recommended values:

5: Standard choice (default)
3: Large datasets/quick evaluation
10: Thorough evaluation/smaller datasets

Trade-offs:

Higher values: More thorough, computationally expensive
Lower values: Faster, potentially higher variance

Must be at least 2 for valid cross-validation.

RandomState

u64

Random seed for fold generation when shuffling. Important for:

Reproducible results
Consistent fold assignments
Benchmark comparisons
Debugging and validation

Set specific value for reproducibility across runs.

Shuffle

bool

true

Whether to shuffle data before splitting into folds. Effects:

true: Randomized fold composition (recommended)
false: Sequential splitting

Enable when:

Data may have ordering
Better fold independence needed

Disable for:

Time series data
Ordered observations

StratifiedKfoldCv

Stratified K-fold cross-validation maintaining class proportions across folds.

Key features:

Preserves class distribution in each fold
Handles imbalanced datasets
Ensures representative splits

Best for:

Classification problems
Imbalanced class distributions
When class proportions matter

Requirements:

Classification tasks only
Sufficient samples per class
Categorical target variable

NSplits

u32

Number of stratified folds. Guidelines: Typical values:

5: Standard for most cases
3: Quick evaluation/large datasets
10: Detailed evaluation/smaller datasets

Considerations:

Must allow sufficient samples per class per fold
Balance between stability and computation time
Consider smallest class size when choosing

RandomState

u64

Seed for reproducible stratified splits. Ensures:

Consistent fold assignments
Reproducible results
Comparable experiments
Systematic validation

Fixed seed guarantees identical stratified splits.

Shuffle

bool

false

Data shuffling before stratified splitting. Impact:

true: Randomizes while maintaining stratification
false: Maintains data order within strata

Use cases:

true: Independent observations
false: Grouped or sequential data

Class proportions maintained regardless of setting.

ShuffleSplitCv

Random permutation cross-validator with independent sampling.

Characteristics:

Random sampling for each split
Independent train/test sets
More flexible than K-fold
Can have overlapping test sets

Advantages:

Control over test size
Fresh splits each iteration
Good for large datasets

Limitations:

Some samples might never be tested
Others might be tested multiple times
No guarantee of complete coverage

NSplits

u32

Number of random splits to perform. Consider: Common values:

5: Standard evaluation
10: More thorough assessment
3: Quick estimates

Trade-offs:

More splits: Better estimation, longer runtime
Fewer splits: Faster, less stable estimates

Balance between computation and stability.

RandomState

u64

Random seed for reproducible shuffling. Controls:

Split randomization
Sample selection
Result reproducibility

Important for:

Debugging
Comparative studies
Result verification

TestSize

f64

0.2

Proportion of samples for test set. Guidelines: Common ratios:

0.2: Standard (80/20 split)
0.25: More validation emphasis
0.1: More training data

Considerations:

Dataset size
Model complexity
Validation requirements

It must be between 0.0 and 1.0.

StratifiedShuffleSplitCv

Stratified random permutation cross-validator combining shuffle-split with stratification.

Features:

Maintains class proportions
Random sampling within strata
Independent splits
Flexible test size

Ideal for:

Imbalanced datasets
Large-scale problems
When class distributions matter
Flexible validation schemes

NSplits

u32

Number of stratified random splits. Guidelines: Recommended values:

5: Standard evaluation
10: Detailed analysis
3: Quick assessment

Consider:

Sample size per class
Computational resources
Stability requirements

RandomState

u64

Seed for reproducible stratified sampling. Ensures:

Consistent class proportions
Reproducible splits
Comparable experiments

Critical for:

Benchmarking
Research studies
Quality assurance

TestSize

f64

0.2

Fraction of samples for stratified test set. Best practices: Common splits:

0.2: Balanced evaluation
0.3: More thorough testing
0.15: Preserve training size

Consider:

Minority class size
Overall dataset size
Validation objectives

It must be between 0.0 and 1.0.

Time Series cross-validator. Provides train/test indices to split time series data samples that are observed at fixed time intervals, in train/test sets. It is a variation of k-fold which returns first k folds as train set and the k + 1th fold as test set. Note that unlike standard cross-validation methods, successive training sets are supersets of those that come before them. Also, it adds all surplus data to the first training partition, which is always used to train the model. Key features:

Maintains temporal dependence
Expanding window approach
Forward-chaining splits
No future data leakage

Use cases:

Sequential data
Financial forecasting
Temporal predictions
Time-dependent patterns

Note: Training sets are supersets of previous iterations.

NSplits

u32

Number of temporal splits. Considerations: Typical values:

5: Standard forward chaining
3: Limited historical data
10: Long time series

Impact:

Affects training window growth
Determines validation points
Influences computational load

MaxTrainSize

u64

Maximum size of training set. Should be strictly less than the number of samples. Applications:

0: Use all available past data
>0: Rolling window of fixed size

Use cases:

Limit historical relevance
Control computational cost
Handle concept drift
Memory constraints

TestSize

u64

Number of samples in each test set. When 0:

Auto-calculated as n_samples/(n_splits+1)
Ensures equal-sized test sets

Considerations:

Forecast horizon
Validation requirements
Available future data

Gap

u64

Number of samples to exclude from the end of each train set before the test set.Gap between train and test sets. Uses:

Avoid data leakage
Model forecast lag
Buffer periods

Common scenarios:

0: Continuous prediction
>0: Forward gap for realistic evaluation
Match business forecasting needs