No packages match

n4m - Portable Partial Least Squares and NIRS Engine

Implements a portable Partial Least Squares (PLS) and Near-Infrared Spectroscopy (NIRS) engine. Provides fit/predict wrappers for the shipped PLS regression solvers (NIPALS, SIMPLS, SVD, kernel, wide-kernel, orthogonal-scores, power, randomized SVD, PCR), variants (sparse SIMPLS, CPPLS, weighted, robust, ridge, continuum, multi-block, GLM, MIR), adaptive AOM-PLS and POP-PLS operator selection, variable-selection methods (SPA, CARS, GA, random frog, stability selection, VIP), diagnostics (Hotelling T2, Q residuals, DModX), and calibration transfer (PDS, DS). The 'C++17' implementation is vendored and compiled from source at install time; no external system libraries are required.

Last updated

cpp

6.27 score 22 stars 1 dependents 33 scripts

pls4all - Partial Least Squares and Chemometrics Engine

A portable Partial Least Squares engine for chemometrics: the slim, PLS-focused distribution carved from the 'nirs4all-methods' library. It ships every method built on the shared PLS core (NIPALS, SIMPLS, SVD, kernel, wide-kernel, orthogonal-scores, power, randomized SVD, PCR): regression variants (sparse SIMPLS, CPPLS, weighted, robust, ridge, continuum, multi-block, GLM, MIR), adaptive AOM-PLS / POP-PLS operator selection, variable-selection methods (SPA, CARS, GA, random frog, stability selection, VIP), PLS diagnostics (Hotelling T2, Q residuals, DModX), and calibration transfer (PDS, DS). The spectroscopy-specific surface (spectral preprocessing, augmentation, sample filters, signal-type detection) lives in the full 'nirs4all-methods' distribution. The same 'C++17' numerical core powers both; here it is vendored and compiled from source at install time, with no external system libraries required.

Last updated

cpp

5.27 score 22 stars 9 scripts

nirs4allformats - Read Near-Infrared Spectroscopy and Spectral File Formats

Thin 'R' binding for the 'nirs4all-formats' near-infrared spectroscopy (NIRS) file-loading core written in 'Rust'. When installed via 'R CMD INSTALL' with 'Cargo' available, the package compiles a native 'extendr' static library from 'src/rust/' and dispatches probe, read, and walk calls directly through 'Rust'. Without 'Cargo' it falls back to invoking the 'nirs4all-formats' command-line interface. This is the complete build: it ships every reader, including the optional large ones ('HDF5'/'netCDF', 'Parquet'/'Arrow', 'MATLAB') on top of the core readers ('JCAMP-DX', 'Galactic SPC', 'Bruker OPUS', 'ASD', 'ENVI', 'CSV', 'Excel', and many vendor ASCII/binary formats). A smaller sibling package 'nirs4allformats.lite' drops only the 'Parquet'/'Arrow' reader for size-sensitive installs.

Last updated

rustcargo

4.23 score 24 stars 6 scripts

nirs4alldatasets - Acquire Curated NIRS Reference Datasets ('nirs4all-datasets')

'R' binding over the 'nirs4all-datasets' C 'ABI' ('n4ds_*'): resolve a dataset id from the distributable catalog index into a neutral descriptor plus version-pinned download contract, fetch the canonical 'Parquet' ('Dataverse' / 'Zenodo' / 'figshare') with SHA-256 verification into a local cache, and re-verify a cached directory offline. JSON crosses the stable C 'ABI'; analysis of the data is left to the host.

Last updated

rustcargo

3.59 score 6 stars 2 scripts 8 downloads

dagml - R Binding for DAG-ML Local Implementations

Provides R bindings for native DAG-ML phase execution, host hyperparameter search, cross-validation, refit, prediction and bundle replay through process adapters. R loss and metric functions are retained in process-local registries keyed by validated implementation descriptors; executable functions are never serialized into graph contracts.

Last updated

3.48 score 4 stars 5 scripts

dagmldata - R Provider Bindings for dag-ml-data

Thin extendr bindings exposing the dag-ml-data in-memory provider (materialize, view, identity, target, feature, collation) to R as JSON in, JSON out. NIRS, ML and scheduling logic live in dag-ml-data / dag-ml.

Last updated

rustcargo

3.45 score 4 stars 1 scripts

nirs4allio - Dataset-Assembly Bridge for the 'nirs4all' Ecosystem

'R' binding over the 'nirs4all-io' C 'ABI' ('n4io_*'): normalize inputs into a canonical 'DatasetSpec', infer a 'DatasetPlan', validate a 'DatasetSpec', and materialize an assembled structural summary. The JSON surface crosses the stable C 'ABI'; the low-level 'n4io_*' functions take and return canonical JSON strings, while the idiomatic 'nio_*' layer accepts native 'R' inputs (a path, a vector of files, or a config list) and returns typed S3 objects with print and as.data.frame methods.

Last updated

rustcargo

3.38 score 4 stars 6 scripts

nirs4all - Composable Near-Infrared Spectroscopy Pipelines for R

Provides an R-facing pipeline API for near-infrared spectroscopy. Composes portable preprocessing (including local and robust normalization, area normalization, detrending, train-fitted MSC and Savitzky-Golay filtering) and partial least squares operations from 'n4m' with explicit R learner controllers, including native sparse PLS-DA and post-SIMPLS group-sparse PLS, optional 'parsnip' and 'mlr3' regression and classification adapters, 'ranger' probability-forest and optional 'xgboost' regression and classification controllers, and optional CPU 'torch' regression and classification controllers. Offers local fit, predict, portable 'n4m' model and bounded trained-pipeline serialization, and a native 'dagml' cross-validation, fixed-candidate selection, refit and replay path for single-node pipelines and optional sequential n4m transform nodes, including fixed-candidate selection and parallel preprocessing branches joined and composed with sequential preprocessing before a model. Native-only n4m refit artifacts are detached into portable N4MM bytes in the DAG bundle; other persisted refit models and transform states remain R sidecars. Both can predict new R samples. Optional 'nirs4allformats' datasets feed local and native paths with numeric or categorical targets while preserving sample and spectral-axis identities. Shared native dense-regression workflows provide out-of-fold selection, resumable hyperparameter search, experiment result views, held-out split conformal calibration and frozen robustness audits. Complete native multimodal predictors preserve aligned raw source schemas and portable fitted state.

Last updated

3.17 score 98 scripts