Options for configuring LOWESS smoothing.
More...
#include <fastlowess.hpp>
Options for configuring LOWESS smoothing.
◆ auto_converge
| double fastlowess::LowessOptions::auto_converge = NAN |
Auto-convergence threshold.
◆ backend
| std::string fastlowess::LowessOptions::backend = "cpu" |
Execution backend: "cpu" (default) or "gpu" (requires the library to be built with the gpu Cargo feature and a Vulkan/Metal/DX12-capable GPU).
◆ boundary_policy
| std::string fastlowess::LowessOptions::boundary_policy = "extend" |
◆ custom_weights
| std::vector<double> fastlowess::LowessOptions::custom_weights |
Per-observation case weights.
When non-empty, must have the same length as the data passed to fit(). All values must be finite and non-negative.
◆ cv
Cross-validation configuration. Empty fractions disable CV.
◆ delta
| double fastlowess::LowessOptions::delta = NAN |
Interpolation threshold (NaN = auto)
◆ fraction
Smoothing fraction (0, 1].
◆ intervals
◆ iterations
| int fastlowess::LowessOptions::iterations = 3 |
◆ missing
| std::string fastlowess::LowessOptions::missing = "error" |
Policy for non-finite (NaN/Inf) values in input data: "error" (default) or "drop".
◆ outputs
| std::vector<std::string> fastlowess::LowessOptions::outputs |
Optional result components: diagnostics, residuals, weights, derivative, se, and sorted.
◆ parallel
| bool fastlowess::LowessOptions::parallel = true |
◆ retain_model
| bool fastlowess::LowessOptions::retain_model = false |
◆ robustness_method
| std::string fastlowess::LowessOptions::robustness_method = "bisquare" |
◆ scaling_method
| std::string fastlowess::LowessOptions::scaling_method = "mad" |
◆ seed
| std::optional<uint64_t> fastlowess::LowessOptions::seed |
◆ weight_function
| std::string fastlowess::LowessOptions::weight_function = "tricube" |
◆ zero_weight_fallback
| std::string fastlowess::LowessOptions::zero_weight_fallback = "use_local_mean" |
The documentation for this struct was generated from the following file: