---

name: ds-dimensionality-reduction compatibility: opencode completeness: 95 content-types:

• code
• guidance
• do-dont
• examples description: '"Provides Reduces data dimensionality using PCA, t-SNE, UMAP, autoencoders and other feature extraction methods for visualization and efficiency"' license: MIT maturity: stable metadata: domain: coding output-format: code related-skills: ds-clustering, ds-community-detection, ds-eda, ds-feature-engineering role: implementation scope: implementation triggers: dimensionality reduction, PCA, t-SNE, UMAP, feature extraction, how do i reduce dimensions archetypes:
  • tactical
  • generation anti_triggers:
  • brainstorming
  • vague ideation
  • code golf
  • over-engineering response_profile: verbosity: low directive_strength: high abstraction_level: operational version: "1.0.0"

---

Dimensionality Reduction

Comprehensive guide to dimensionality reduction in machine learning and data science workflows.

When to Use This Skill

• Solving real-world unsupervised learning problems
• Building machine learning pipelines with dimensionality reduction
• Implementing best practices for dimensionality reduction
• Optimizing model performance using dimensionality reduction techniques
• Learning industry-standard approaches to dimensionality reduction

When NOT to Use This Skill

• When using pre-built libraries without understanding underlying concepts
• For toy problems that don't require dimensionality reduction rigor
• When domain expertise in specific problem requires different approach
• If your problem doesn't require the complexity this skill provides

Purpose and Key Concepts

Dimensionality Reduction is a critical component of the machine learning workflow. This skill covers:

1. Theoretical foundations — Mathematical principles and statistical concepts
2. Practical implementation — Working code examples and patterns
3. Common pitfalls — Mistakes to avoid and how to recover from them
4. Best practices — Industry-standard approaches and optimization techniques

Core Workflow

1. Understand the problem — Clearly define what you're solving for
2. Select approach — Choose the right technique for your data and constraints
3. Implement solution — Write clean, tested code following best practices
4. Validate results — Verify your implementation with tests and validation
5. Optimize performance — Improve efficiency and accuracy incrementally

Implementation Patterns

Pattern 1: Basic Dimensionality Reduction

import numpy as np
import pandas as pd
from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler
from sklearn.datasets import load_iris

def basic_dimensionality_reduction(data: pd.DataFrame, n_components: int = 2) -> pd.DataFrame:
    """Apply PCA for basic dimensionality reduction with proper scaling."""
    if data.empty:
        raise ValueError("Input DataFrame cannot be empty")
    
    scaler = StandardScaler()
    scaled_data = scaler.fit_transform(data)
    
    pca = PCA(n_components=n_components)
    reduced_data = pca.fit_transform(scaled_data)
    
    component_names = [f"PC{i+1}" for i in range(n_components)]
    reduced_df = pd.DataFrame(reduced_data, columns=component_names, index=data.index)
    
    explained_variance = pca.explained_variance_ratio_
    print(f"Explained variance ratio: {explained_variance}")
    print(f"Total variance explained: {sum(explained_variance):.2%}")
    
    return reduced_df

# Example usage
if __name__ == "__main__":
    iris = load_iris()
    df = pd.DataFrame(iris.data, columns=iris.feature_names)
    result = basic_dimensionality_reduction(df, n_components=2)
    print(result.head())

Pattern 2: Production-Ready Dimensionality Reduction

import logging
import numpy as np
import pandas as pd
from typing import Any, Dict, Optional
from sklearn.decomposition import PCA
from sklearn.preprocessing import StandardScaler
import umap

logger = logging.getLogger(__name__)

class ProductionDimensionalityReduction:
    """Production-grade dimensionality reduction pipeline with method switching."""
    
    def __init__(self, method: str = "pca", n_components: int = 2, random_state: int = 42):
        self.method = method
        self.n_components = n_components
        self.random_state = random_state
        self.scaler = StandardScaler()
        self.reducer = None
        self.is_fitted = False

    def _initialize_reducer(self) -> None:
        if self.method == "pca":
            self.reducer = PCA(n_components=self.n_components, random_state=self.random_state)
        elif self.method == "umap":
            self.reducer = umap.UMAP(n_components=self.n_components, random_state=self.random_state)
        else:
            raise ValueError(f"Unsupported method: {self.method}")

    def execute(self, data: pd.DataFrame) -> Dict[str, Any]:
        """Execute dimensionality reduction on input data with full validation."""
        if data.empty:
            raise ValueError("Input data cannot be empty")
        
        self._initialize_reducer()
        scaled_data = self.scaler.fit_transform(data)
        reduced_data = self.reducer.fit_transform(scaled_data)
        self.is_fitted = True
        
        result = {
            "reduced_data": reduced_data
            "method": self.method
            "n_components": self.n_components
            "explained_variance": getattr(self.reducer, "explained_variance_ratio_", None)
            "shape": reduced_data.shape
        }
        logger.info(f"Successfully reduced {data.shape[1]} dimensions to {self.n_components}")
        return result

# Example usage
if __name__ == "__main__":
    df = pd.DataFrame(np.random.randn(100, 10), columns=[f"feat_{i}" for i in range(10)])
    pipeline = ProductionDimensionalityReduction(method="pca", n_components=2)
    output = pipeline.execute(df)
    print(f"Reduced shape: {output['shape']}")

Best Practices

• ✅ Always validate your implementation on test data
• ✅ Document your assumptions and methodology
• ✅ Use version control for reproducibility
• ✅ Monitor performance metrics in production
• ✅ Periodically review and update your approach
• ✅ Test with edge cases and outliers
• ✅ Log all significant operations for debugging

Common Pitfalls

| Pitfall | Problem | Solution | |

---

---

Constraints

MUST DO

• Validate all data preprocessing steps are fit-only on training data, never on validation or test sets
• Implement reproducible pipelines with fixed random seeds and deterministic operations where possible
• Report model performance with confidence intervals via bootstrapping or cross-validation across multiple runs
• Log all experiments with parameters, metrics, and artifacts using MLflow or equivalent tracking system

MUST NOT DO

• Do not evaluate a model on the same data used for training — always hold out a proper test set
• Avoid overfitting to the validation set by limiting hyperparameter search iterations
• Never use features that can only be computed at inference time (look-ahead bias)
• Do not report single-run accuracy without statistical significance testing or error bars

Live References

Authoritative documentation links for this skill's domain. The model follows markdown links at load time to resolve external references and inline content.

• Scikit-learn Decomposition Methods
• PCA — Scikit-learn docs
• t-SNE Visualization (Jupyter Notebook by Christopher Olah)
• UMAP — Manifold Learning
• Feature Extraction & Selection (Kaggle Learn)
• PCA vs t-SNE Comparison (Distill.pub)