Utilities

Submodules

taurex.util.emission module

Functions related to computing emission spectrums

black_body

black_body_numba

black_body_numba_II

black_body_numexpr(lamb, temp)

black_body_numpy(lamb, temp)

integrate_emission_layer(dtau, layer_tau, mu, BB)

integrate_emission_numba

taurex.util.math module

Optimized Math functions used in taurex

class OnlineVariance

Bases: object

USes the M2 algorithm to compute the variance in a streaming fashion

combine_variance(averages, variance, counts)

parallelVariance()

reset()

property sampleVariance

update(value, weight=1.0)

property variance

compute_rayleigh_cross_section(wngrid, n, n_air=2.6867805e+25, king=1.0)

intepr_bilin

intepr_bilin_double(x11, x12, x21, x22, T, Tmin, Tmax, P, Pmin, Pmax)

intepr_bilin_numba

intepr_bilin_numba_II

intepr_bilin_numexpr(x11, x12, x21, x22, T, Tmin, Tmax, P, Pmin, Pmax)

intepr_bilin_old(x11, x12, x21, x22, T, Tmin, Tmax, P, Pmin, Pmax)

interp_exp_and_lin(x11, x12, x21, x22, T, Tmin, Tmax, P, Pmin, Pmax)

2D interpolation

Applies linear interpolation across P and e interpolation across T between Pmin,Pmax and Tmin,Tmax

Parameters

• x11 (array) – Array corresponding to Pmin,Tmin

• x12 (array) – Array corresponding to Pmin,Tmax

• x21 (array) – Array corresponding to Pmax,Tmin

• x22 (array) – Array corresponding to Pmax,Tmax

• T (float) – Coordinate to exp interpolate to

• Tmin (float) – Nearest known T coordinate where Tmin < T

• Tmax (float) – Nearest known T coordinate where T < Tmax

• P (float) – Coordinate to linear interpolate to

• Pmin (float) – Nearest known P coordinate where Pmin < P

• Pmax (float) – Nearest known P coordinate where P < Tmax

interp_exp_and_lin_broken

interp_exp_and_lin_numpy(x11, x12, x21, x22, T, Tmin, Tmax, P, Pmin, Pmax)

2D interpolation

Applies linear interpolation across P and e interpolation across T between Pmin,Pmax and Tmin,Tmax

Parameters

• x11 (array) – Array corresponding to Pmin,Tmin

• x12 (array) – Array corresponding to Pmin,Tmax

• x21 (array) – Array corresponding to Pmax,Tmin

• x22 (array) – Array corresponding to Pmax,Tmax

• T (float) – Coordinate to exp interpolate to

• Tmin (float) – Nearest known T coordinate where Tmin < T

• Tmax (float) – Nearest known T coordinate where T < Tmax

• P (float) – Coordinate to linear interpolate to

• Pmin (float) – Nearest known P coordinate where Pmin < P

• Pmax (float) – Nearest known P coordinate where P < Tmax

interp_exp_numba

interp_exp_numpy(x11, x12, T, Tmin, Tmax)

interp_exp_only(x11, x12, T, Tmin, Tmax)

interp_lin_numba

interp_lin_numpy(x11, x12, P, Pmin, Pmax)

interp_lin_only

test_nan(val)

taurex.util.util module

General utility functions

bindown(original_bin, original_data, new_bin, last_point=None)

This method quickly bins down by taking the mean. The numpy histogram function is exploited to do this quickly

Parameters

• original_bin (numpy.array) – The original bins for the that we want to bin down

• original_data (numpy.array) – The associated data that will be averaged along the new bins

• new_bin (numpy.array) – The new binnings we want to use (must have less points than the original)

Returns

Binned mean of original_data

Return type

array

calculate_weight(chem)

class_for_name(class_name)

class_from_keyword(keyword, class_filter=None)

clip_native_to_wngrid(native_grid, wngrid)

compute_bin_edges(wngrid)

compute_dz(altitude)

conversion_factor(from_unit, to_unit)

create_grid_res(resolution, wave_min, wave_max)

decode_string_array(f)

Helper to decode strings from hdf5

ensure_string_utf8(val)

find_closest_pair(arr, value) -> (<class 'int'>, <class 'int'>)

Will find the indices that lie to the left and right of the value

arr[left] <= value <= arr[right]

If the value is less than the array minimum then it will always return left=0 and right=1

If the value is above the maximum

Parameters

• arr (array) – Array to search, must be sorted

• value (float) – Value to find in array

Returns

• left (int)

• right (int)

get_molecular_weight(gasname)

For a given molecule return the molecular weight in atomic units

Parameters

gasname (str) – Name of molecule

Returns

molecular weight in amu or 0 if not found

Return type

float

has_duplicates(arr)

loadtxt2d(intext)

Wraps loadtext and either returns a 2d array or 1d array

merge_elements(elem1, elem2, factor=1)

molecule_texlabel(gasname)

For a given molecule return its latex form

Parameters

gasname (str) – Name of molecule

Returns

Latex form of the molecule or just the passed name if not found

Return type

str

movingaverage(a, n=3)

Computes moving average

Parameters

• a (array) – Array to compute average

• n (int) – Averaging window

Returns

Resultant array

Return type

array

quantile_corner(x, q, weights=None)

• Taken from corner.py

__author__ = “Dan Foreman-Mackey (danfm@nyu.edu)” __copyright__ = “Copyright 2013-2015 Daniel Foreman-Mackey”

Like numpy.percentile, but:

• Values of q are quantiles [0., 1.] rather than percentiles [0., 100.]

• scalar q not supported (q must be iterable)

• optional weights on x

Parameters

• x (array) – Input array or object that can be converted to an array.

• q (array or float) – Percentile or sequence of percentiles to compute, which must be between 0 and 1 inclusive.

• weights (array or float , optional) – Weights on x

Returns

percentile

Return type

scalar or ndarray

random_int_iter(total, fraction)

read_error_into_dict(l, d)

Reads the error into dict?

read_error_line(l)

Reads line?

read_table(txt, d=None, title=None)

Yeah whatever i give up

recursively_save_dict_contents_to_output(output, dic)

Will recursive write a dictionary into output.

Parameters

• output (Output or OutputGroup) – Group (or root) in output file to write to

• dic (dict) – Dictionary we want to write

sanitize_molecule_string(molecule)

Cleans a molecule string to match up with molecule naming in TauREx3.

e.g:

H2O -> H2O

1H2-16O -> H2O

Parameters

molecule (str) – Molecule to sanitize

Returns

Sanitized name

Return type

str

split_molecule_elements(molecule=None, tokens=None)

store_thing(output, key, item)

tokenize_molecule(molecule)

weighted_avg_and_std(values, weights, axis=None)

Computes weight average and standard deviation

Parameters

• values (array) – Input array

• weights (array) – Must be same shape as values

axisint , optional

axis to perform weighting

wnwidth_to_wlwidth(wngrid, wnwidth)

Module contents

Common functions that are used and are quite helpful