GPU Backend
Run the batch LOWESS fit on the GPU via wgpu (Vulkan/Metal/DX12, GLES on Android) instead of the CPU.
Overview
The batch Lowess type can execute on a GPU-accelerated backend powered by wgpu. It reimplements almost the entire LOWESS pipeline — local regression fitting, robustness iterations, interval bounds, and cross-validation — as WGSL compute shaders, so all anchor points are fit in parallel instead of one at a time on CPU cores.
This is worth enabling for high-throughput processing of large datasets (roughly 10k+ points); for smaller inputs the CPU backend (optionally with parallel(true)) is typically faster once you account for GPU dispatch overhead. See BENCHMARKS.md for crossover points measured on real hardware.
|
Batch only. GPU support applies to the batch |
GPU support is opt-in and not included in a default build of fastlowess-java — download a prebuilt GPU-enabled build via the one-time installer below, or build the native library from source with the gpu Cargo feature.
Supported Features
-
Weight Functions: All standard kernels (
tricube,epanechnikov,gaussian,uniform,biweight,triangle,cosine). -
Robustness Methods:
bisquare,huber, andtalwarrobustness weighting. -
Scaling Methods: Residual scaling using
mad(Median Absolute Deviation),mar(Median Absolute Residual), andmean(Mean Absolute Residual). -
Interval Bounds: GPU-native computation of standard errors, confidence intervals, and prediction intervals.
-
Optimization:
-
Parallel Fitting: Local regression for all anchor points is computed in parallel.
-
Robustness Loops: Iterative weight updates and convergence checks occur entirely on the GPU.
-
Distance-based Skipping: Support for the
deltaparameter to accelerate smoothing on dense grids. -
Validation: GPU-accelerated
kfoldandloocvcross-validation.
Feature Comparison
| Feature | CPU | GPU | Notes |
|---|---|---|---|
Batch fitting |
✅ |
✅ |
|
Streaming/Online |
✅ |
❌ |
GPU optimized for static batch data |
All weight/robustness/scaling methods |
✅ |
✅ |
|
Confidence/prediction intervals |
✅ |
✅ |
|
Cross-validation (k-fold, LOOCV) |
✅ |
✅ |
|
Unsorted inputs and sorted-output option |
✅ |
✅ |
GPU results preserve Batch output ordering |
Custom per-observation weights |
✅ |
✅ |
Applied in GPU fit and CV candidate kernels |
Each compute pipeline now binds only its own resources, so adapter limits are based on the largest individual GPU pass rather than the union of all shader bindings.
Checking Availability
Before requesting backend("gpu"), check whether the currently loaded native library was built with GPU support:
if (!FastLowess.gpuEnabled()) {
throw new IllegalStateException("this build of fastlowess_java was not compiled with the gpu feature");
}
Requesting the GPU backend when it isn’t available throws a clear RuntimeException from the Lowess constructor, rather than a raw crash.
Installing GPU Support
Download and install a prebuilt GPU-enabled fastlowess_java native library for this platform from the gpu-builds release (built by .github/workflows/release-gpu.yml) — a single perpetual release holding GPU artifacts for every version, so individual version release pages stay uncluttered; the source version is embedded in each asset’s filename instead:
FastLowess.installGpu(); // prompts for confirmation, then downloads
Non-interactively: FastLowess.installGpu(true). The library is saved to ~/.fastlowess/gpu/ under the standard platform library name. A running JVM cannot swap an already-loaded native library — restart the JVM afterwards, pointing NativeBridge at the downloaded build via -Dfastlowess.native.dir=<dir printed above>.
Building from source with the gpu Cargo feature is always available as an alternative:
cargo build -p fastlowess-java --release --features gpu
Since NativeBridge loads the shared library by path (via the fastlowess.native.dir system property), simply point it at the GPU-enabled build — no application rebuild is required.
Usage
Once GPU support is available, request it by setting backend on the batch options:
Options options = Options.builder()
.backend("gpu")
.intervals(IntervalsOptions.builder().confidence(0.95).build())
.build();
try (Lowess model = new Lowess(options)) {
Result result = model.fit(x, y);
}
If GPU support isn’t available, the Lowess constructor throws a RuntimeException pointing at how to build a GPU-enabled library, rather than crashing.