Utility functions

_trapzx2(x1, x2, y1, y2)

Twice the area under the line between (x1, y1) and (x2, y2). Used in trapezoidal integration

_vander(x, n)

Calculate the the Vandermonde matrix. See https://en.wikipedia.org/wiki/Vandermonde_matrix

banded_inds(row, span, row_len)

Calculate the bounds of the filled indices for row row of a banded matrix of span span

bounds_generator!(bounds, xs)

Getting the bounds of each element in xs assuming that they fully span the domain

bounds_generator!(xs)

Getting the bounds of each element in xs assuming that they fully span the domain

clip_vector!(vec; max=Inf, min=-Inf)

Set values in vec above max to max and below min to min

copy_dict!(to, from)

Copy all entries in from to to

est_∇(f, inputs; dif=1e-7, ignore_0_inputs=false)

Estimate the gradient of f at inputs using finite differences.

find_modes(data; amount=3)

Return the indices of local maxima of a data array

flatten_ranges(ranges)

Returns a range from the largest first element of those in ranges to the smallest last element of those in ranges

fwhm_2_σ(fwhm)

Convert full-width at half-maximum to σ

general_lst_sq(dm, data, Σ)

Solve a weighted linear system of equations. See https://en.wikipedia.org/wiki/Generalizedleastsquares#Method_outline

general_lst_sq_f(data, Σ, order; x=eachindex(data))

Get a polynomial of order order fit to data

int2ind(a, x)

Find the index where x can be found in a

make_template(matrix, σ²; default=1., use_mean=false, kwargs...)

Reduce matrix to its median (or weighted mean) and clip the result

make_template(matrix; use_mean=false, kwargs...)

Reduce matrix to its median (or mean) and clip the result

multiple_append!(a, b...)

Generalized version of the Julia's append!() function

observation_night_inds(times_in_days)

Find the indices for each observing night

ordinary_lst_sq(dm, data)

Solve a linear system of equations. See https://en.wikipedia.org/wiki/Ordinaryleastsquares#Matrix/vector_formulation

ordinary_lst_sq_f(data, order; x=eachindex(data))

Get a polynomial of order order fit to data

oversamp_interp(lo_x, hi_x, x, y)

Interpolating by getting the average value of y from lo_x to hi_x

parse_args(ind, type, default)

Retrieve ARGS[ind] as type type if it exists. Otherwise, return default

searchsortednearest(a, x; lower=false)

Find the index of the value closest to x in a (which is a sorted vector)

self_cor(a; set_diag=true)

Get the correlation matrix between the rows in a

trapz_small(x, y)

Trapezoidal integration of y over x. Shamelessly modified from https://github.com/dextorious/NumericalIntegration.jl/blob/master/src/NumericalIntegration.jl. See https://en.wikipedia.org/wiki/Trapezoidal_rule

trapz_small(lo_x, hi_x, x, y)

Trapezoidal integration of y over x from lo_x to hi_x. See https://en.wikipedia.org/wiki/Trapezoidal_rule

vector_zero(θ)

Get a zero version of θ

wavenumber_to_Å(wn)

Convert wn (in 1/cm) to wavelength (in Å)

weighted_mean(x, σ²; default=0., kwargs...)

Calculates the weighted mean of x given σ²

Å_to_wavenumber(λ)

Convert λ (in Å) to wave number (in 1/cm)