R and CPU Comparison | CPU-Parallel versus GPU | End-to-End GPU Fit | Host/Device Transfer Overhead

Overview | Comparing fitted values | Why defaults differ | What this package adds

Overview | Confidence Intervals | Prediction Intervals | Both Intervals | Confidence Levels | Standard Errors | Availability

How Custom Weights Work | When to Use Custom Weights | Custom Weights vs. Robustness Iterations | Basic Usage | Suppress a Known Outlier | Emphasize Important Points | Propagate Measurement Uncertainty | Combined with Robustness Iterations | Validation Rules | See Also

Quick Decision Guide | Batch Adapter | When to Use | Example | Streaming Adapter | Parameters | Merge Strategies | Online Adapter | Update Modes | Feature Comparison | Next Steps

Overview | K-Fold Cross-Validation | Leave-One-Out (LOOCV) | Seeded Randomization | Comparison | CV Metrics | Interpreting Results | Availability | Best Practices

Basic Smoothing | With Confidence Intervals | Handling Outliers | Streaming Mode | Online Mode | Plotting Results | Next Steps

Overview | Methylation Profile Smoothing | The Challenge | Solution | ChIP-seq Signal Smoothing | Application | Large Genome Coverage (Streaming) | Best Practices for Genomic Data | See Also

Overview | Basic Trend Extraction | Seasonal Decomposition | Irregular Time Grids | Forecasting with Prediction Intervals | Handling Missing Data | Multi-Scale Analysis | Gene Expression Time Course | Choosing Fraction for Time Series | See Also

Overview | Average | Take First | Take Last | Weighted Average | Choosing a Strategy | Choosing Chunk Size and Overlap

Overview | 1D — Standard (Default) | 2D — Spatial Surface | 3D and Higher | Distance Metrics for Multivariate Data | Surface Mode for Multivariate Data

Overview | Degree 0 — Local Constant | Degree 1 — Local Linear (Default) | Degree 2 — Local Quadratic | Degree 3 — Local Cubic | Degree 4 — Local Quartic | Degree and Boundary Fallback | Choosing the Right Degree | Higher Degree Effects | Surface Mode

How Robustness Works | Robustness Methods | Bisquare (Default) | Huber | Talwar | Comparison | Detecting Outliers | Scale Estimation | Auto-Convergence

Overview | Supported Features on GPU | Checking GPU Availability | Installing GPU Support | Using the GPU Backend | Performance Guidelines | Fallback Behaviour

Quick Decision Guide | Batch Adapter | When to Use | Example | Streaming Adapter | Parameters | Merge Strategies | Online Adapter | Update Modes | Feature Comparison

Overview | Confidence Intervals | Prediction Intervals | Both Intervals | Confidence Levels | Bootstrap Intervals | Standard Errors | Availability

Basic Smoothing | With Confidence Intervals | Handling Outliers | Streaming Mode | Next Steps

Overview | Basic Trend Extraction | Detrending | Forecasting with Prediction Intervals | Handling Missing Data | Multi-Scale Analysis | Gene Expression Time Course | Choosing Fraction for Time Series | See Also

Overview | Reproducing stats::lowess() | Why the defaults differ | What this package adds

Overview | K-Fold Cross-Validation | Leave-One-Out (LOOCV) | Seeded Randomization | Comparison | CV Metrics | Interpreting Results | Availability | Best Practices

Overview | Weighted Average (Default) | Average | Take First | Take Last | Choosing Chunk Size and Overlap | Choosing a Strategy

Overview | Methylation Profile Smoothing | The Challenge | Solution | ChIP-seq Signal Smoothing | Application | Large Genome Coverage (Streaming) | Best Practices for Genomic Data | See Also

How Robustness Works | Robustness Methods | Bisquare (Default) | Huber | Talwar | Comparison | Detecting Outliers | Scale Estimation | Auto-Convergence

Overview | Online Mode: Point-by-Point | Sensor Data Example | Streaming Mode: Chunk Processing | Log File Processing | Real-Time Dashboard Example | Choosing Parameters | Online Mode | Streaming Mode | Performance Considerations | See Also

Overview | Extend (Default) | Reflect | Zero | No Boundary | Comparing Policies | Choosing a Policy

What is LOESS? | How It Works | The Fraction Parameter | Robustness Iterations | Confidence vs Prediction Intervals | Execution Modes | Quick Decision Guide | Key Advantages | Next Steps

Overview | MAD — Median Absolute Deviation (Default) | MAR — Median Absolute Residual | Mean — Mean Absolute Residual | Choosing a Scaling Method | Comparing Scaling Methods

Overview | Online Mode: Point-by-Point | Sensor Data Example | Accumulating Results | Streaming Mode: Chunk Processing | Log File Processing | Real-Time Dashboard Example | Choosing Parameters | Online Mode | Streaming Mode | Performance Considerations | See Also

Overview | Available Kernels | Tricube (Default) | Epanechnikov | Gaussian | Biweight | Cosine | Triangle | Uniform | Choosing a Kernel

How Custom Weights Work | When to Use Custom Weights | Custom Weights vs. Robustness Iterations | Basic Usage | Suppress a Known Outlier | Emphasize Important Points | Propagate Measurement Uncertainty | Combined with Robustness Iterations | Validation Rules | See Also

What is LOWESS? | How It Works | The Fraction Parameter | Robustness Iterations | Confidence vs Prediction Intervals | Execution Modes | Quick Decision Guide | Key Advantages | Next Steps

Overview | MAD — Median Absolute Deviation (Default) | MAR — Median Absolute Residual | Mean — Mean Absolute Residual | Choosing a Scaling Method

Overview | Available Kernels | Tricube (Default) | Epanechnikov | Gaussian | Biweight | Cosine | Triangle | Uniform | Choosing a Kernel

Overview | Extend (Default) | Reflect | Zero | No Boundary Padding | Comparing Policies

From R-universe (recommended) | From conda-forge | From Source (GitHub) | System Requirements | Verifying Installation

From R-universe (recommended) | From conda-forge | From Source (GitHub) | System Requirements | Verifying Installation