Kneighbors / Classifier Layer

K-Nearest Neighbors Classifier - Classifier implementing the k-nearest neighbors vote.

Mathematical form: $\overset{y}{^} (x) = ar g y max i \in N_{k} (x) \sum w_{i} I (y_{i} = y)$ where:

$N_{k} (x)$ is the k nearest neighbors of point x
$w_{i}$ is the weight of the i-th neighbor
$I ()$ is the indicator function

Key characteristics:

Instance-based learning
No training phase
Local decision making
Distance-based classification
Lazy learning algorithm

Common applications:

Pattern recognition
Recommendation systems
Anomaly detection
Image classification
Medical diagnosis

Outputs:

Predicted Table: Input data with predictions
Validation Results: Cross-validation metrics
Test Metric: Test set performance
ROC Curve Data: ROC analysis information
Confusion Matrix: Classification breakdown
Feature Importances: Distance-based importance

Note: Performance heavily depends on feature scaling and distance metric choice

Table

Predicted Table

Validation Results

Test Metric

ROC Curve Data

Confusion Matrix

Feature Importances

SelectFeatures

[column, ...]

Feature columns for k-Nearest Neighbors classification:

Requirements:

Numerical features
No missing values
Finite values only
Distance-compatible features

Preprocessing guidelines:

Scaling (critical):
- StandardScaler: When normal distribution
- MinMaxScaler: When bounded range needed
- RobustScaler: When outliers present
Feature engineering:
- Dimensionality reduction
- Handle categorical variables
- Create meaningful distances
- Remove redundant features
Distance considerations:
- Feature relevance to distance
- Feature interactions
- Curse of dimensionality
- Distance metric validity
Quality checks:
- Feature distributions
- Outlier detection
- Scale compatibility
- Distance correlation

If empty, uses all numeric columns except target.

Note: Feature scaling is crucial for distance-based algorithms

SelectTarget

column

Target column for k-NN classification:

Requirements:

Categorical labels
No missing values
At least two classes
Properly encoded

Local prediction characteristics:

Class probabilities from local density
Sensitive to local class distribution
Adapts to decision boundary shape
Non-parametric estimation

Class considerations:

Local class balance
Spatial class distribution
Class overlap regions
Decision boundary complexity

Quality checks:

Validate label consistency
Check class distributions
Monitor local densities
Verify label encoding

Params

oneof

DefaultParams

Optimized default configuration for k-Nearest Neighbors:

Default settings:

n_neighbors: 5 (balanced local influence)
weights: uniform (simple voting)
algorithm: auto (adaptive selection)
leaf_size: 30 (balanced tree structure)
power: 2 (Euclidean distance)

Best suited for:

Medium-sized datasets
Balanced classes
Scaled features
Initial exploration

Computational complexity:

Query time: $O (lo g n)$ for tree-based
Space: $O (n)$ for storage

CustomParams

Fine-grained control over k-Nearest Neighbors parameters:

Parameter categories:

Model structure (n_neighbors)
Distance computation (weights, power)
Algorithm optimization (algorithm, leaf_size)

Performance impact:

Time complexity
Memory usage
Prediction accuracy
Decision boundary smoothness

NNeighbors

u32

Number of neighbors for classification:

Selection guide:

Small k: More local, sensitive to noise
Large k: Smoother boundaries, biased
Rule of thumb: $k \approx n$

Trade-offs:

Bias vs. variance
Noise sensitivity
Computation time

Weights

enum

uniform

Neighbor weighting scheme for prediction:

Weight calculation:

Uniform: $w_{i} = 1$
Distance: $w_{i} = \frac{1}{d ( x , x _{i} )}$

Impact:

Affects neighbor influence
Changes decision boundaries
Modifies prediction confidence

uniform ~

Equal weights for all neighbors:

Properties:

Simple majority voting
Distance-independent
More robust to noise
Stable predictions

distance ~

Inverse distance weighting:

Properties:

Closer neighbors more important
Smooth decision boundaries
Distance-sensitive
Better for continuous spaces

Algorithm

enum

auto

Algorithm for computing nearest neighbors:

Selection criteria:

Data dimensionality
Sample size
Distance metric
Memory constraints

auto ~

Automatic algorithm selection:

Chooses based on:

Dataset size
Feature dimensionality
Metric type
Available memory

ball_tree ~

Ball Tree algorithm:

Best for:

High dimensions (n > 3)
Complex distance metrics
Variable density data
Memory efficiency needed

kd_tree ~

KD Tree algorithm:

Best for:

Low dimensions (n ≤ 3)
Euclidean distance
Uniform density data
Fast queries needed

brute ~

Brute force search:

Best for:

Small datasets
Very high dimensions
Custom metrics
When exact NN needed

LeafSize

u32

Tree algorithm leaf size parameter:

Trade-offs:

Small: Faster queries, more memory
Large: Slower queries, less memory

Optimal value depends on:

Dataset size
Available memory
Query frequency

Power

f64

Minkowski distance power parameter:

Common values:

p=1: Manhattan distance
p=2: Euclidean distance (default)
p=∞: Chebyshev distance

Choose based on:

Feature space properties
Domain knowledge
Performance requirements

GridSearch

Hyperparameter optimization for k-Nearest Neighbors:

Search strategy:

Model structure tuning
Distance metric optimization
Algorithm selection

Best practices:

Start with n_neighbors search
Consider data characteristics
Monitor computational resources

NNeighbors

[u32, ...]

Neighbor count values to evaluate:

Common grids:

Basic: [3, 5, 7]
Extended: [3, 5, 7, 9, 11]
Logarithmic: [1, 2, 4, 8, 16]

Consider dataset size when choosing range

Weights

[enum, ...]

uniform

Neighbor weighting scheme for prediction:

Weight calculation:

Uniform: $w_{i} = 1$
Distance: $w_{i} = \frac{1}{d ( x , x _{i} )}$

Impact:

Affects neighbor influence
Changes decision boundaries
Modifies prediction confidence

uniform ~

Equal weights for all neighbors:

Properties:

Simple majority voting
Distance-independent
More robust to noise
Stable predictions

distance ~

Inverse distance weighting:

Properties:

Closer neighbors more important
Smooth decision boundaries
Distance-sensitive
Better for continuous spaces

Algorithm

[enum, ...]

auto

Algorithm for computing nearest neighbors:

Selection criteria:

Data dimensionality
Sample size
Distance metric
Memory constraints

auto ~

Automatic algorithm selection:

Chooses based on:

Dataset size
Feature dimensionality
Metric type
Available memory

ball_tree ~

Ball Tree algorithm:

Best for:

High dimensions (n > 3)
Complex distance metrics
Variable density data
Memory efficiency needed

kd_tree ~

KD Tree algorithm:

Best for:

Low dimensions (n ≤ 3)
Euclidean distance
Uniform density data
Fast queries needed

brute ~

Brute force search:

Best for:

Small datasets
Very high dimensions
Custom metrics
When exact NN needed

LeafSize

[u32, ...]

Leaf size values to evaluate:

Typical ranges:

Small: [10, 20, 30]
Medium: [30, 50, 70]
Large: [50, 100, 150]

Impact on tree construction and query speed

Power

[f64, ...]

Distance metric powers to evaluate:

Common values:

[1.0]: Manhattan distance
[2.0]: Euclidean distance
[1.0, 2.0, 3.0]: Compare metrics

Consider feature space geometry

RefitScore

enum

Accuracy

Performance metric for model evaluation:

Selection criteria:

Default: Model's built-in scoring
Accuracy: Overall correctness
BalancedAccuracy: For imbalanced data
LogLoss: Probability quality
RocAuc: Threshold-independent

Choose based on:

Class distribution
Problem requirements
Prediction type needed

Default ~

Uses estimator's built-in scoring method:

For Bernoulli NB:

Returns accuracy score
Equal weight to all samples
Fast computation

Best for:

Quick evaluation
Balanced datasets
Initial testing

Accuracy ~

Standard classification accuracy score:

Formula: $Accuracy = \frac{TP + TN}{TP + TN + FP + FN}$

Properties:

Range: [0, 1]
Perfect score: 1.0
Baseline: max(class proportions)

Best for:

Balanced classes
Equal error costs
Simple evaluation

BalancedAccuracy ~

Class-weighted accuracy score:

Formula: $Balanced Accuracy = \frac{1}{n _{c l a sses}} i = 1 \sum n_{c l a sses} \frac{T P _{i}}{T P _{i} + F N _{i}}$

Properties:

Adjusts for class imbalance
Range: [0, 1]
Random baseline: 0.5

Best for:

Imbalanced datasets
When minority classes matter
Uneven class distributions

LogLoss ~

Logarithmic loss (cross-entropy):

Formula: $- \frac{1}{N} i = 1 \sum N j = 1 \sum M y_{ij} lo g (p_{ij})$

Properties:

Penalizes confident mistakes
Range: [0, ∞)
Perfect score: 0.0

Best for:

Probability calibration
When confidence matters
Probabilistic predictions

RocAuc ~

Area Under Receiver Operating Characteristic Curve:

Properties:

Threshold-independent
Range: [0, 1]
Random baseline: 0.5
Perfect score: 1.0

Best for:

Binary classification
Threshold tuning
Ranking evaluation
Imbalanced datasets

Note: For multiclass, computes average ROC AUC

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