galapy.internal.utils

Functions

`cat_to_dict`(infile[, id_field, err_field, ...])	Converts an ASCII (e.g. Topcat-like) catalogue into a 2-levels dictionary.
`filter_strings`(inlist, fields)	Function for filtering a list of strings (uses `fnmatch`)
`find_nearest`(array, value)	Function finding the indexes of array closer to the values in value.
`func_scalar_or_array`(var, function, *args, ...)	A function applying some function to a scalar or an ndarray.
`get_credible_interval`(samples, idcentre[, ...])	Find lower and upper limits in sample array containing given percent value around position.
`get_nested`(d, kl)
`now_string`()	Generates a string with the minute the function is called
`poly_N`(xx, coeff)	Method for computing N-order polynomia.
`powerlaw_exp_cutoff`(El, gamma, Ecut)	Returns a powerlaw with exponential cut-off of the form
`quantile_weighted`(values, quantiles[, ...])	Very close to numpy.percentile, but supports weights.
`set_nested`(d, kl, v)	Recursive function for setting values in a nested dictionary.
`shorten_string`(string[, maxlength, splitchar])	Function for shortening a string, splitting it in components.
`trap_int`(xx, yy)	Trapezoid integration
`unwrap_items`(d)	Generator yielding the items of a nested dictionary
`unwrap_keys`(d)	Generator yielding the keys of a nested dictionary
`unwrap_values`(d)	Generator yielding the values of a nested dictionary

Classes

FlagVal(value, flag)

galapy.internal.utils.trap_int(xx, yy)

Trapezoid integration

Parameters:

xx (array-like) – x-domain grid
yy (array-like) – y-domain grid

Returns:

the integral along the whole x-domain

Return type:

float

galapy.internal.utils.find_nearest(array, value)

Function finding the indexes of array closer to the values in value.

Parameters:

array (array-like)
value (array-like or scalar)

Returns:

list of indexes of elements in array closer to values in value.

Return type:

numpy-array

Warning

Uniqueness of elements in the returned array is not guaranteed.

galapy.internal.utils.powerlaw_exp_cutoff(El, gamma, Ecut)

Returns a powerlaw with exponential cut-off of the form

\[y = E_\lambda^{-\gamma+3} e^{E_\lambda/E_\text{cut}}\]

Parameters:

El (scalar or array-like) – the x-values
gamma (scalar) – the spectral index
Ecut (scalar) – the cut-off x-scale

Return type:

scalar or array-like

galapy.internal.utils.poly_N(xx, coeff)

Method for computing N-order polynomia. The order of the polynomium is set by the length of the coeff list.

The method computes

\[y = \sum_0^N c_i * x^{N-i}\]

Parameters:

xx (scalar or array-like) – the x-values
coeff (list) – the values of the coefficients of the polynomium. The list must be ordered, with the first element corresponding to the coefficient multiplying the highest power of the x-variable and the last coefficient being the scalar coefficient

Returns:

the y-values

Return type:

scalar or array-like

galapy.internal.utils.unwrap_keys(d)

Generator yielding the keys of a nested dictionary

Parameters:: d (dictionary) – A nested dictionary (whathever object inheriting from ‘dict’ or collections.abc.MutableMapping or similar)
Yields:: list of keys

Examples

>>> d = { 'a' : 1, 'b' : { 'c' : 2, 'd' : 3 } }
>>> for i in unwrap_keys( d ) : print( i )
['a']
['b', 'c']
['b', 'd']

galapy.internal.utils.unwrap_values(d)

Generator yielding the values of a nested dictionary

Parameters:: d (dictionary) – A nested dictionary (whathever object inheriting from ‘dict’ or collections.abc.MutableMapping or similar)
Yields:: values

Examples

>>> d = { 'a' : 1, 'b' : { 'c' : 2, 'd' : 3 } }
>>> for i in unwrap_values( d ) : print( i )
1
2
3

galapy.internal.utils.unwrap_items(d)

Generator yielding the items of a nested dictionary

Parameters:: d (dictionary) – A nested dictionary (whathever object inheriting from ‘dict’ or collections.abc.MutableMapping or similar)
Yields:: items – a 2d tuple in which the first element is the list of nested keys and the second is the associated value

Examples

>>> d = { 'a' : 1, 'b' : { 'c' : 2, 'd' : 3 } }
>>> for i in unwrap_items( d ) : print( i )
(['a'], 1)
(['b', 'c'], 2)
(['b', 'd'], 3)

galapy.internal.utils.set_nested(d, kl, v)

Recursive function for setting values in a nested dictionary.

Parameters:

d (dictionary) – the i/o dictionary
kl (list) – a list of keywords to traverse the nested dictionary the first element is the highest position in the hierarchy of the nested dictionary
v (whatever) – the value to set at corresponding nested sequence of keywords

Return type:

None

Examples

>>> d = {}
>>> kl1 = ['first', 'second', 'third']
>>> kl2 = ['a', 'b', 'c']
>>> set_nested(d,kl1, 3.)
>>> set_nested(d,kl2, True)
>>> d
{'first': {'second': {'third': 3.0}}, 'a': {'b': True}}

If we want to add a new key-value pair to an already existing level:

>>> kl3 = ['first', 'second', 'fourth']
>>> set_nested(d, kl3, 4.)
>>> d
{'first': {'second': {'third': 3.0, 'fourth': 4.0}},'a': {'b': True}}

galapy.internal.utils.func_scalar_or_array(var, function, *args, **kwargs)

A function applying some function to a scalar or an ndarray.

Parameters:

var (scalar or ndarray) – The argument of the function
function (callable) – A function taking a scalar as input
args (sequence) – A list or tuple of additional arguments to be passed to function
kwargs (dict) – A dictionary of additional keyword arguments to be passed to function

Returns:

Depends on the shape of var: (i) if var is a scalar it is the result of function(var); (ii) if var is an ndarray it is the result of array([function(v) for v in var])

Return type:

scalar or ndarray

galapy.internal.utils.quantile_weighted(values, quantiles, weights=None, values_sorted=False, old_style=False, axis=-1)

Very close to numpy.percentile, but supports weights. (partially stolen from stackoverflow: https://stackoverflow.com/questions/21844024/weighted-percentile-using-numpy)

Parameters:

values (numpy.array) – with data
quantiles (array-like) – quantiles needed (NOTE: quantiles should be in [0, 1]!)
sample_weight (array-like) – same length as values
values_sorted (bool) – if True, then will avoid sorting of initial array
old_style (bool) – if True, will correct output to be consistent with numpy.percentile.
axis (int)

Returns:

computed quantiles.

Return type:

numpy.array

galapy.internal.utils.get_credible_interval(samples, idcentre, percent=0.68, weights=None)

Find lower and upper limits in sample array containing given percent value around position.

It splits the samples in 2 half-intervals: [lower, samples[idcentre]) and [samples[idcentre], upper) that are set to contain 0.5*percent of all the weighted samples. If one of the two half-intervals does not have enough samples, it is considered un-defined. When this happens, the function consider it instead an upper/lower limit on the sampled parameter and returns instead (None, upper-limit) or (lower-limit, None).

Parameters:

samples (ndarray or iterable) – the weighted samples over which to compute the estimator
idcentre (integer) – index in the samples iterable around which to compute the interval
percent (float) – probability enclosed by the interval
weights – (Optional, default = None) weights of the samples, if not None this must have the same size of the samples argument

Returns:

lower, upper – [lower-, upper-) limits of the interval if the interval is defined. If it is not defined, the function will return either an upper limit or a lower limit, respectively.

Return type:

scalar floats

galapy.internal.utils.now_string(): Generates a string with the minute the function is called

galapy.internal.utils.filter_strings(inlist, fields)

Function for filtering a list of strings (uses fnmatch)

Parameters:

inlist (sequence of str) – A list of strings to filter
fields (str or sequence of str) – Either a single string or a sequence of strings. Also accepts wildcards (e.g. which_params = 'sfh*' will show all the entries in inlist that contain the sub-string sfh).

Return type:

str or sequence of str

Raises:

ValueError –

galapy.internal.utils.shorten_string(string, maxlength=10, splitchar='.')

Function for shortening a string, splitting it in components.

Parameters:

string (str) – Input string
maxlength (int) – (default = 10) Maximum number of characters in the final string
splitchar (str) – (default = ‘.’) The character over which perform the splitting.

Returns:

str
The shortened string with length <= maxlength+len(splitchar)

galapy.internal.utils.cat_to_dict(infile, id_field='id', err_field='_err', meta_fields=[], skip_fields=[])

Converts an ASCII (e.g. Topcat-like) catalogue into a 2-levels dictionary. The 1st order dictionary contains a 2nd order dictionary for each entry in the catalogue. Each 2nd order dictionary contains data and meta-data about the named entry.

Parameters:

infile (str) – Path to the ASCII file
id_field (str) – (Optional, default=’id’) header name of the field containing the sources’ ID
err_field (str) – (Optional, default=’_err’) the sub-string identifying fields containing error measurements
meta_fields (str or sequence of str) – (Optional, default empty list) which fields to consider meta-data Either a single string or a sequence of strings. Also accepts wildcards (e.g. meta_fields = 'R*' will consider all the fields in header that start with R as meta-data).
skip_fields (str or sequence of str) – (Optional, default empty list) which fields to skip Either a single string or a sequence of strings. Also accepts wildcards (e.g. skip_fields = 'R*' will ignore all the fields in header that start with R).

Return type:

dict