fosanalysis.utils.misc Namespace Reference

Contains miscellaneous standalone functions. More...

Functions
np.array	datetime_to_timestamp (np.array datetime_array)
	Converts an array of datetime entries to an array of Unix timestamps.
tuple	find_closest_value (np.array arr, float v)
	Returns index and value in arr, that is closest to the given v.
np.array	last_finite_index (np.array arr, int axis=-1)
	Returns an array with the indices of the most recent finite entry when traversing the arr along the specified axis.
np.array	nan_diff (np.array arr, int axis=-1)
	Calculate the difference to the previous finite entry.
np.array	nan_diff_1d (np.array arr)
	Calculate the difference to the previous finite entry.
tuple	next_finite_neighbor (np.array array, int index, bool to_left, int recurse=0)
	Find the next finite neighbor of the entry array[index].
	np_to_python (data)
	Convert the given data to a Python built-in type.
np.array	timestamp_to_datetime (np.array timestamp_array)
	Converts an array of Unix timestamps to an array of datetime entries.

Detailed Description

Contains miscellaneous standalone functions.

Author

Bertram Richter

Date

2023

Function Documentation

datetime_to_timestamp()

np.array fosanalysis.utils.misc.datetime_to_timestamp

(

np.array

datetime_array

)

Converts an array of datetime entries to an array of Unix timestamps.

Parameters

datetime_array

Array of datetime objects.

Returns

Returns an array of Unix timestamps.

Definition at line 163 of file misc.py.

find_closest_value()

tuple fosanalysis.utils.misc.find_closest_value	(	np.array	arr,
		float	v )

Returns index and value in arr, that is closest to the given v.

In case of equal distance of v to both neighbors, the smaller one is chosen.

Parameters

arr	Array like (1D) of values in ascending order.
v	The target value, to which the distance should be minimized.

Returns

(<index>, <entry>)

Definition at line 11 of file misc.py.

last_finite_index()

np.array fosanalysis.utils.misc.last_finite_index	(	np.array	arr,
		int	axis = -1 )

Returns an array with the indices of the most recent finite entry when traversing the arr along the specified axis.

The returned array has the same shape as arr.

Example:

>>> arr = np.array([1.,2.,"nan", "inf", 5], dtype=float)
array([1., 2., nan, inf, 5.])
>>> last_finite_index(arr)
array([0, 1, 1, 1, 4])

The first element is assumed to be a finite value. All indices before the first finite entry will be 0.

>>> arr = np.array(["nan","nan", "inf", 5], dtype=float)
array([nan, nan, inf, 5.])
>>> last_finite_index(arr)
array([0, 0, 0, 3])

Parameters

arr	Array like.
axis	Axis along which to apply the indexing. Defaults to the last axis.

Definition at line 34 of file misc.py.

Here is the caller graph for this function:

nan_diff()

np.array fosanalysis.utils.misc.nan_diff	(	np.array	arr,
		int	axis = -1 )

Calculate the difference to the previous finite entry.

This is similar to np.diff(), but skipping NaN or inf entries. This function is a wrapper around nan_diff_1d().

Example:

>>> arr = np.array([1.,2.,"nan", "inf", 5], dtype=float)
array([1., 2., nan, inf, 5.])
>>> nan_diff(arr)
array([1., nan, -inf, 3.])

Parameters

arr	Array like.
axis	Axis along which to calculate the incremental difference. Defaults to the last axis.

Definition at line 93 of file misc.py.

nan_diff_1d()

np.array fosanalysis.utils.misc.nan_diff_1d

(

np.array

arr

)

Calculate the difference to the previous finite entry.

This is similar to np.diff(), but skipping NaN or inf entries.

Example:

>>> arr = np.array([1.,2.,"nan", "inf", 5], dtype=float)
array([1., 2., nan, inf, 5.])
>>> nan_diff_1d(arr)
array([1., nan, -inf, 3.])

Parameters

arr	Array like, needs to be 1D.

Definition at line 71 of file misc.py.

Here is the call graph for this function:

next_finite_neighbor()

tuple fosanalysis.utils.misc.next_finite_neighbor	(	np.array	array,
		int	index,
		bool	to_left,
		int	recurse = 0 )

Find the next finite neighbor of the entry array[index].

An entry <entry> is finite, if np.isfinite(<entry>) == True.

Parameters

array	Array, on which the search is carried out.
index	Position in the array, where the search is started.
to_left	True, if a neighbor to the left of the starting index should be found, False for a neighbor to the right.
recurse	Number of recursions, that are done. Examples: 0: direct neighbors of the starting index 1: neighbors of the neighbors 2: neighbors of the neighbors' neighbors and so on.

Returns

Tuple like (<index>, <entry>). If no finite value could be found before reaching the end of array (None, None) is returned.

Definition at line 114 of file misc.py.

Here is the call graph for this function:

Here is the caller graph for this function:

np_to_python()

fosanalysis.utils.misc.np_to_python

(

data

)

Convert the given data to a Python built-in type.

This function should be used, when type-checking. Iterables are recursively converted into list and instances of np.scalar are converted into standard data types using the method np.item().

Definition at line 150 of file misc.py.

Here is the call graph for this function:

Here is the caller graph for this function:

timestamp_to_datetime()

np.array fosanalysis.utils.misc.timestamp_to_datetime

(

np.array

timestamp_array

)

Converts an array of Unix timestamps to an array of datetime entries.

Parameters

timestamp_array

Array of floats, representing Unix timestamps.

Returns

Returns an array of datetime.datetime objects.

Definition at line 171 of file misc.py.