Basic Usage

Fit a spectrum with the default BIC model selection:

from gaussFitSpec import fit_spectrum, plot_fit, read_spectrum

velocity, spectrum, spectrum_err = read_spectrum("examples/example_spectra.txt")

result = fit_spectrum(
    velocity,
    spectrum,
    spectrum_err,
    name="spectra1",
    method="bic",
    max_components=8,
)

print(result.components)
result.to_csv("examples/example_components.csv")

plot_fit(
    velocity,
    spectrum,
    spectrum_err,
    result,
output_file="examples/example_fit.png",
)

Fitting Options

Use min_fwhm and max_fwhm to constrain component widths:

result = fit_spectrum(
    velocity,
    spectrum,
    spectrum_err,
    min_fwhm=2.0,
    max_fwhm=40.0,
)

Fit Result

The returned SpectrumFitResult contains:

components: A pandas DataFrame with one row per Gaussian component.
best_model: The total fitted Gaussian model evaluated on the input velocity grid.
individual_components: A 2D array containing each Gaussian component evaluated on the input velocity grid.
residual: spectrum - best_model.
method: The selected model-selection method.
n_components: The number of selected Gaussian components.
fit_statistics: A dictionary with values such as chi2, reduced_chi2, aic, and bic.

Component Table

The component table includes:

name
component_index
amplitude
amplitude_err
velocity
velocity_err
fwhm
fwhm_err