API¶

Chemometrics: A chemometrics package for Python¶

chemometrics is a Python module providing chemometric functionality in the Python ecosystem for scientific computing and machine learning (numpy, scipy, matplotlib, scikit-learn).

Visualisation¶

`plot_colored_series`(Y[, x, reference])	Plot lines colored by position or reference
`plot_svd`(D[, n_comp, n_eigenvalues])	Plot SVD-matrices in three subplots.

Preprocessing¶

`Emsc`([p_order, background, normalize, ...])	Perform extended multiplicative scatter correction (EMSC)
`AsymWhittaker`(penalty[, constraint_order, ...])	Background correction X with an asymmetric Whittaker filter
`Whittaker`([penalty, constraint_order, deriv])	Smooth with a Whittaker filter

Decomposition¶

`PCA`([n_components, copy, whiten, ...])	Principal component analysis with added chemometric functionality
`fit_pca`(X[, pipeline, cv_object, max_lv])	Auto-calibrate PCA model and generate analytical plots

Regression¶

`PLSRegression`([n_components, max_iter, tol, ...])	PLS regression with added chemometric functionality
`fit_pls`(X, Y[, pipeline, cv_object, max_lv])	Auto-calibrate PLS model and generate analytical plots
`IHM`(features, peak_parameters[, bl_order, ...])	Indirect Hard Modeling (IHM) without linear regression
`IHMRegression`(features, peak_parameters[, ...])	Indirect Hard Modeling (IHM) of spectra with OLS prediction

Multivariate curve resolution¶

The chemometrics.mcr module provides Multivariate Curve Resolution - Alternating Least Squares (MCR-ALS) functionality

`mcr.McrAR`([c_regr, st_regr, fit_kwargs, ...])	Multivariate Curve Resolution - Alternating Regression
`mcr.constraint`	McrAR constraints
`mcr.regressor`	Built-in least squares / regression methods.
`mcr.metric`	Metrics used in `chemometrics.mcr`

Artificial data generation¶

`generate_background`(n_wl[, rel_lengthscale, ...])	Generate dummy background.
`generate_data`([n_wl, n_samples, n_conc, noise])	Generate artificial spectroscopic XY data without background
`generate_spectra`(n_wl, n_band, bandwidth)	Generate a dummy spectra with n_band
`pseudo_voigt_spectra`(x, parameter)	Generates vector based on pseudo-Voigt profiles