Package 'rfastloess'

Title: High-Performance LOESS Smoothing for R
Description: Provides high-performance LOESS smoothing (Locally Estimated Scatterplot Smoothing) using a 'Rust' backend. Supports various weight functions, robustness iterations, streaming/online processing, cross-validation, and uncertainty quantification. Applicable to genomic data, time series, and general-purpose smoothing tasks.
Authors: Amir Valizadeh [aut, cre, fnd] (ORCID: <https://orcid.org/0000-0001-5983-8527>)
Maintainer: Amir Valizadeh <[email protected]>
License: MIT + file LICENSE | Apache License (== 2.0)
Version: 0.9.0
Built: 2026-07-12 23:39:36 UTC
Source: https://github.com/thisisamirv/loess-project

Help Index

rfastloess: High-performance LOESS Smoothing for R

Description

A high-performance LOESS (Locally Estimated Scatterplot Smoothing) implementation built on the Rust fastLoess crate.

For comprehensive documentation, see: https://github.com/thisisamirv/loess-project/tree/main/docs

Main Classes

Documentation

For comprehensive documentation, tutorials, and API reference, see: https://loess.readthedocs.io/

Author(s)

Maintainer: Amir Valizadeh [email protected] (ORCID) [funder]

Authors:

See Also

Useful links:

Examples

# Basic smoothing
x <- seq(1, 10, length.out = 100)
y <- sin(x) + rnorm(100, sd = 0.2)
model <- Loess(fraction = 0.3)
result <- model$fit(x, y)
plot(x, y)
lines(result$x, result$y, col = "red", lwd = 2)

LOESS Batch Smoothing

Description

Create a stateful LOESS model for batch smoothing.

Usage

Loess(
    fraction = 0.67,
    iterations = 3L,
    weight_function = "tricube",
    robustness_method = "bisquare",
    scaling_method = "mad",
    boundary_policy = "extend",
    confidence_intervals = NULL,
    prediction_intervals = NULL,
    return_diagnostics = FALSE,
    return_residuals = FALSE,
    return_robustness_weights = FALSE,
    zero_weight_fallback = "use_local_mean",
    auto_converge = NULL,
    cv_fractions = NULL,
    cv_method = "kfold",
    cv_k = 5L,
    parallel = TRUE,
    degree = "linear",
    dimensions = 1L,
    distance_metric = "normalized",
    surface_mode = "interpolation",
    return_se = FALSE,
    weighted_metric_weights = NULL,
    cell = NULL,
    interpolation_vertices = NULL,
    boundary_degree_fallback = NULL,
    cv_seed = NULL
)

Arguments

fraction

Smoothing fraction (between 0 and 1).
iterations

Number of robustness iterations (non-negative integer). Default: 3.
weight_function

Kernel weight function. One of "tricube" (default), "gaussian", "uniform", "cosine", "epanechnikov", "biweight", "triangle".
robustness_method

Outlier downweighting method: "bisquare" (default), "huber", or "talwar".
scaling_method

Residual scale estimation for robustness weights: "mad" (default), "mar", or "mean".
boundary_policy

Boundary handling strategy: "extend" (default), "reflect", "zero", or "noboundary".
confidence_intervals

Confidence level for confidence intervals (e.g., 0.95). NULL (default) disables confidence intervals.
prediction_intervals

Confidence level for prediction intervals (e.g., 0.95). NULL (default) disables prediction intervals.
return_diagnostics

Logical; if TRUE, return fit-quality metrics (RMSE, MAE, R-squared, AIC, etc.). Default: FALSE.
return_residuals

Logical; if TRUE, return residuals in the result. Default: FALSE.
return_robustness_weights

Logical; if TRUE, return per-point robustness weights. Default: FALSE.
zero_weight_fallback

Fallback policy when all robustness weights drop to zero: "use_local_mean" (default), "return_original", or "return_none".
auto_converge

Convergence tolerance for early stopping of robustness iterations. NULL (default) disables early stopping.
cv_fractions

Numeric vector of candidate fractions for cross-validation. NULL (default) disables CV.
cv_method

Cross-validation method: "kfold" (default) or "loocv".
cv_k

Number of folds for k-fold CV. Default: 5.
parallel

Logical; enable parallel processing. Default: TRUE.
degree

Local polynomial degree: "constant", "linear" (default), "quadratic", "cubic", or "quartic".
dimensions

Number of predictor dimensions. Default: 1.
distance_metric

Distance metric for neighbourhood computation: "normalized" (default), "euclidean", "manhattan", "chebyshev", "minkowski", or "weighted". Use "minkowski:p" to set a custom p value.
surface_mode

Surface evaluation mode: "interpolation" (default) or "direct".
return_se

Logical; if TRUE, compute hat-matrix statistics (effective degrees of freedom, leverage, standard errors). Default: FALSE.
weighted_metric_weights

Numeric vector of per-dimension weights used when distance_metric = "weighted". Length must equal dimensions. NULL (default) uses equal weights.
cell

Cell size tuning parameter for the interpolation grid. NULL (default) uses the library default.
interpolation_vertices

Number of vertices in the interpolation grid. NULL (default) uses the library default.
boundary_degree_fallback

Logical; if TRUE, fall back to lower polynomial degree at boundaries when fitting at the requested degree fails. NULL (default) uses the library default.
cv_seed

Integer seed for the cross-validation random number generator. NULL (default) uses a random seed.

Value

A Loess object.

Examples

x <- seq(0, 10, length.out = 100)
y <- sin(x) + rnorm(100, 0, 0.1)
model <- Loess(fraction = 0.2)
result <- model$fit(x, y)
plot(x, y)
lines(x, result$y, col = "red")

Nullable Value Wrapper

Description

Wraps a value to be passed to Rust as an Option.

Usage

Nullable(x)

Arguments

x

Value to wrap or NULL.

Value

The value itself. This is a helper for rextendr conversion.

Examples

Nullable(5)
Nullable(NULL)

LOESS Online Smoothing

Description

Create a stateful LOESS model for real-time online data.

Usage

OnlineLoess(
    fraction = 0.67,
    window_capacity = 1000L,
    min_points = 3L,
    iterations = 3L,
    weight_function = "tricube",
    robustness_method = "bisquare",
    scaling_method = "mad",
    boundary_policy = "extend",
    update_mode = "full",
    auto_converge = NULL,
    return_robustness_weights = FALSE,
    return_diagnostics = FALSE,
    return_residuals = FALSE,
    zero_weight_fallback = "use_local_mean",
    parallel = FALSE,
    degree = "linear",
    dimensions = 1L,
    distance_metric = "normalized",
    surface_mode = "interpolation",
    return_se = FALSE,
    confidence_intervals = NULL,
    prediction_intervals = NULL,
    weighted_metric_weights = NULL,
    cell = NULL,
    interpolation_vertices = NULL,
    boundary_degree_fallback = NULL
)

Arguments

fraction

Smoothing fraction (between 0 and 1).
window_capacity

Maximum number of points kept in the sliding window.
min_points

Minimum number of points required before smoothing begins.
iterations

Number of robustness iterations (non-negative integer). Default: 3.
weight_function

Kernel weight function. One of "tricube" (default), "gaussian", "uniform", "cosine", "epanechnikov", "biweight", "triangle".
robustness_method

Outlier downweighting method: "bisquare" (default), "huber", or "talwar".
scaling_method

Residual scale estimation for robustness weights: "mad" (default), "mar", or "mean".
boundary_policy

Boundary handling strategy: "extend" (default), "reflect", "zero", or "noboundary".
update_mode

Window update strategy: "full" (default) re-smooths all window points after each addition, "incremental" updates only the newest point.
auto_converge

Convergence tolerance for early stopping of robustness iterations. NULL (default) disables early stopping.
return_robustness_weights

Logical; if TRUE, return per-point robustness weights. Default: FALSE.
return_diagnostics

Logical; if TRUE, return fit-quality metrics (RMSE, MAE, R-squared, AIC, etc.). Default: FALSE.
return_residuals

Logical; if TRUE, return residuals in the result. Default: FALSE.
zero_weight_fallback

Fallback policy when all robustness weights drop to zero: "use_local_mean" (default), "return_original", or "return_none".
parallel

Logical; enable parallel processing. Default: TRUE.
degree

Local polynomial degree: "constant", "linear" (default), "quadratic", "cubic", or "quartic".
dimensions

Number of predictor dimensions. Default: 1.
distance_metric

Distance metric for neighbourhood computation: "normalized" (default), "euclidean", "manhattan", "chebyshev", "minkowski", or "weighted". Use "minkowski:p" to set a custom p value.
surface_mode

Surface evaluation mode: "interpolation" (default) or "direct".
return_se

Logical; if TRUE, compute hat-matrix statistics (effective degrees of freedom, leverage, standard errors). Default: FALSE.
confidence_intervals

Confidence level for confidence intervals (e.g., 0.95). NULL (default) disables confidence intervals.
prediction_intervals

Confidence level for prediction intervals (e.g., 0.95). NULL (default) disables prediction intervals.
weighted_metric_weights

Numeric vector of per-dimension weights used when distance_metric = "weighted". Length must equal dimensions. NULL (default) uses equal weights.
cell

Cell size tuning parameter for the interpolation grid. NULL (default) uses the library default.
interpolation_vertices

Number of vertices in the interpolation grid. NULL (default) uses the library default.
boundary_degree_fallback

Logical; if TRUE, fall back to lower polynomial degree at boundaries when fitting at the requested degree fails. NULL (default) uses the library default.

Value

An OnlineLoess object.

Examples

model <- OnlineLoess(fraction = 0.2, window_capacity = 20)
x <- 1:50
y <- sin(x * 0.1) + rnorm(50, 0, 0.1)
for (i in seq_along(x)) {
    result <- model$add_point(x[i], y[i])
    if (!is.null(result)) cat("smoothed:", result$smoothed, "\n")
}

Plot Loess Result

Description

Plot Loess Result

Usage

## S3 method for class 'LoessResult'
plot(x, main = "LOESS Fit", ...)

Arguments

x

A LoessResult object.
main

Plot title.
...

Additional arguments passed to plot() and lines().

Value

The input object x, invisibly.

Examples

x <- seq(0, 10, length.out = 100)
y <- sin(x) + rnorm(100, 0, 0.1)
model <- Loess(fraction = 0.2)
res <- model$fit(x, y)
plot(res)

Print Loess Model

Description

Print Loess Model

Usage

## S3 method for class 'Loess'
print(x, ...)

Arguments

x

A Loess object.
...

Additional arguments (ignored).

Value

The input object x, invisibly.

Examples

model <- Loess(fraction = 0.3)
print(model)

Print Loess Result

Description

Print Loess Result

Usage

## S3 method for class 'LoessResult'
print(x, ...)

Arguments

x

A LoessResult object.
...

Additional arguments (ignored).

Value

The input object x, invisibly.

Examples

x <- seq(0, 10, length.out = 50)
y <- sin(x) + rnorm(50, 0, 0.1)
model <- Loess(fraction = 0.3)
result <- model$fit(x, y)
print(result)

Print OnlineLoess Model

Description

Print OnlineLoess Model

Usage

## S3 method for class 'OnlineLoess'
print(x, ...)

Arguments

x

An OnlineLoess object.
...

Additional arguments.

Value

The input object x, invisibly.

Examples

model <- OnlineLoess(fraction = 0.2, window_capacity = 20L)
print(model)

Print StreamingLoess Model

Description

Print StreamingLoess Model

Usage

## S3 method for class 'StreamingLoess'
print(x, ...)

Arguments

x

A StreamingLoess object.
...

Additional arguments.

Value

The input object x, invisibly.

Examples

model <- StreamingLoess(fraction = 0.3, chunk_size = 50L)
print(model)

LOESS Streaming Smoothing

Description

Create a stateful LOESS model for streaming data.

Usage

StreamingLoess(
    fraction = 0.67,
    chunk_size = 5000L,
    overlap = NULL,
    iterations = 3L,
    weight_function = "tricube",
    robustness_method = "bisquare",
    scaling_method = "mad",
    boundary_policy = "extend",
    auto_converge = NULL,
    return_diagnostics = FALSE,
    return_residuals = FALSE,
    return_robustness_weights = FALSE,
    zero_weight_fallback = "use_local_mean",
    merge_strategy = "weighted_average",
    parallel = TRUE,
    degree = "linear",
    dimensions = 1L,
    distance_metric = "normalized",
    surface_mode = "interpolation",
    return_se = FALSE,
    confidence_intervals = NULL,
    prediction_intervals = NULL,
    weighted_metric_weights = NULL,
    cell = NULL,
    interpolation_vertices = NULL,
    boundary_degree_fallback = NULL
)

Arguments

fraction

Smoothing fraction (between 0 and 1).
chunk_size

Number of data points per processing chunk.
overlap

Number of overlapping points between consecutive chunks.
iterations

Number of robustness iterations (non-negative integer). Default: 3.
weight_function

Kernel weight function. One of "tricube" (default), "gaussian", "uniform", "cosine", "epanechnikov", "biweight", "triangle".
robustness_method

Outlier downweighting method: "bisquare" (default), "huber", or "talwar".
scaling_method

Residual scale estimation for robustness weights: "mad" (default), "mar", or "mean".
boundary_policy

Boundary handling strategy: "extend" (default), "reflect", "zero", or "noboundary".
auto_converge

Convergence tolerance for early stopping of robustness iterations. NULL (default) disables early stopping.
return_diagnostics

Logical; if TRUE, return fit-quality metrics (RMSE, MAE, R-squared, AIC, etc.). Default: FALSE.
return_residuals

Logical; if TRUE, return residuals in the result. Default: FALSE.
return_robustness_weights

Logical; if TRUE, return per-point robustness weights. Default: FALSE.
zero_weight_fallback

Fallback policy when all robustness weights drop to zero: "use_local_mean" (default), "return_original", or "return_none".
merge_strategy

Strategy for reconciling overlapping chunk regions: "weighted_average" (default), "average", "take_first", or "take_last".
parallel

Logical; enable parallel processing. Default: TRUE.
degree

Local polynomial degree: "constant", "linear" (default), "quadratic", "cubic", or "quartic".
dimensions

Number of predictor dimensions. Default: 1.
distance_metric

Distance metric for neighbourhood computation: "normalized" (default), "euclidean", "manhattan", "chebyshev", "minkowski", or "weighted". Use "minkowski:p" to set a custom p value.
surface_mode

Surface evaluation mode: "interpolation" (default) or "direct".
return_se

Logical; if TRUE, compute hat-matrix statistics (effective degrees of freedom, leverage, standard errors). Default: FALSE.
confidence_intervals

Confidence level for confidence intervals (e.g., 0.95). NULL (default) disables confidence intervals.
prediction_intervals

Confidence level for prediction intervals (e.g., 0.95). NULL (default) disables prediction intervals.
weighted_metric_weights

Numeric vector of per-dimension weights used when distance_metric = "weighted". Length must equal dimensions. NULL (default) uses equal weights.
cell

Cell size tuning parameter for the interpolation grid. NULL (default) uses the library default.
interpolation_vertices

Number of vertices in the interpolation grid. NULL (default) uses the library default.
boundary_degree_fallback

Logical; if TRUE, fall back to lower polynomial degree at boundaries when fitting at the requested degree fails. NULL (default) uses the library default.

Value

A StreamingLoess object.

Examples

x <- seq(0, 10, length.out = 100)
y <- sin(x) + rnorm(100, 0, 0.1)
model <- StreamingLoess(fraction = 0.2, chunk_size = 50)
res1 <- model$process_chunk(x[1:50], y[1:50])
res2 <- model$process_chunk(x[51:100], y[51:100])
final <- model$finalize()