|
1
|
|
|
# Copyright (c) 2008-2015 MetPy Developers. |
|
2
|
|
|
# Distributed under the terms of the BSD 3-Clause License. |
|
3
|
|
|
# SPDX-License-Identifier: BSD-3-Clause |
|
4
|
|
|
"""Tests for `calc.tools` module.""" |
|
5
|
|
|
|
|
6
|
|
|
import numpy as np |
|
7
|
|
|
import numpy.ma as ma |
|
8
|
|
|
import pytest |
|
9
|
|
|
|
|
10
|
|
|
from metpy.calc import (find_intersections, interpolate_nans, log_interp, |
|
11
|
|
|
nearest_intersection_idx, reduce_point_density, resample_nn_1d) |
|
12
|
|
|
from metpy.calc.tools import _next_non_masked_element, delete_masked_points |
|
13
|
|
|
from metpy.testing import assert_array_almost_equal, assert_array_equal |
|
14
|
|
|
from metpy.units import units |
|
15
|
|
|
|
|
16
|
|
|
|
|
17
|
|
|
def test_resample_nn(): |
|
18
|
|
|
"""Test 1d nearest neighbor functionality.""" |
|
19
|
|
|
a = np.arange(5.) |
|
20
|
|
|
b = np.array([2, 3.8]) |
|
21
|
|
|
truth = np.array([2, 4]) |
|
22
|
|
|
|
|
23
|
|
|
assert_array_equal(truth, resample_nn_1d(a, b)) |
|
24
|
|
|
|
|
25
|
|
|
|
|
26
|
|
|
def test_nearest_intersection_idx(): |
|
27
|
|
|
"""Test nearest index to intersection functionality.""" |
|
28
|
|
|
x = np.linspace(5, 30, 17) |
|
29
|
|
|
y1 = 3 * x**2 |
|
30
|
|
|
y2 = 100 * x - 650 |
|
31
|
|
|
truth = np.array([2, 12]) |
|
32
|
|
|
|
|
33
|
|
|
assert_array_equal(truth, nearest_intersection_idx(y1, y2)) |
|
34
|
|
|
|
|
35
|
|
|
|
|
36
|
|
|
@pytest.mark.parametrize('direction, expected', [ |
|
37
|
|
|
('all', np.array([[8.88, 24.44], [238.84, 1794.53]])), |
|
38
|
|
|
('increasing', np.array([[24.44], [1794.53]])), |
|
39
|
|
|
('decreasing', np.array([[8.88], [238.84]])) |
|
40
|
|
|
]) |
|
41
|
|
|
def test_find_intersections(direction, expected): |
|
42
|
|
|
"""Test finding the intersection of two curves functionality.""" |
|
43
|
|
|
x = np.linspace(5, 30, 17) |
|
44
|
|
|
y1 = 3 * x**2 |
|
45
|
|
|
y2 = 100 * x - 650 |
|
46
|
|
|
# Note: Truth is what we will get with this sampling, not the mathematical intersection |
|
47
|
|
|
assert_array_almost_equal(expected, find_intersections(x, y1, y2, direction=direction), 2) |
|
48
|
|
|
|
|
49
|
|
|
|
|
50
|
|
|
def test_find_intersections_no_intersections(): |
|
51
|
|
|
"""Test finding the intersection of two curves with no intersections.""" |
|
52
|
|
|
x = np.linspace(5, 30, 17) |
|
53
|
|
|
y1 = 3 * x + 0 |
|
54
|
|
|
y2 = 5 * x + 5 |
|
55
|
|
|
# Note: Truth is what we will get with this sampling, not the mathematical intersection |
|
56
|
|
|
truth = np.array([[], |
|
57
|
|
|
[]]) |
|
58
|
|
|
assert_array_equal(truth, find_intersections(x, y1, y2)) |
|
59
|
|
|
|
|
60
|
|
|
|
|
61
|
|
|
def test_find_intersections_invalid_direction(): |
|
62
|
|
|
"""Test exception if an invalid direction is given.""" |
|
63
|
|
|
x = np.linspace(5, 30, 17) |
|
64
|
|
|
y1 = 3 * x ** 2 |
|
65
|
|
|
y2 = 100 * x - 650 |
|
66
|
|
|
with pytest.raises(ValueError): |
|
67
|
|
|
find_intersections(x, y1, y2, direction='increaing') |
|
68
|
|
|
|
|
69
|
|
|
|
|
70
|
|
|
def test_interpolate_nan_linear(): |
|
71
|
|
|
"""Test linear interpolation of arrays with NaNs in the y-coordinate.""" |
|
72
|
|
|
x = np.linspace(0, 20, 15) |
|
73
|
|
|
y = 5 * x + 3 |
|
74
|
|
|
nan_indexes = [1, 5, 11, 12] |
|
75
|
|
|
y_with_nan = y.copy() |
|
76
|
|
|
y_with_nan[nan_indexes] = np.nan |
|
77
|
|
|
assert_array_almost_equal(y, interpolate_nans(x, y_with_nan), 2) |
|
78
|
|
|
|
|
79
|
|
|
|
|
80
|
|
|
def test_interpolate_nan_log(): |
|
81
|
|
|
"""Test log interpolation of arrays with NaNs in the y-coordinate.""" |
|
82
|
|
|
x = np.logspace(1, 5, 15) |
|
83
|
|
|
y = 5 * np.log(x) + 3 |
|
84
|
|
|
nan_indexes = [1, 5, 11, 12] |
|
85
|
|
|
y_with_nan = y.copy() |
|
86
|
|
|
y_with_nan[nan_indexes] = np.nan |
|
87
|
|
|
assert_array_almost_equal(y, interpolate_nans(x, y_with_nan, kind='log'), 2) |
|
88
|
|
|
|
|
89
|
|
|
|
|
90
|
|
|
def test_interpolate_nan_invalid(): |
|
91
|
|
|
"""Test log interpolation with invalid parameter.""" |
|
92
|
|
|
x = np.logspace(1, 5, 15) |
|
93
|
|
|
y = 5 * np.log(x) + 3 |
|
94
|
|
|
with pytest.raises(ValueError): |
|
95
|
|
|
interpolate_nans(x, y, kind='loog') |
|
96
|
|
|
|
|
97
|
|
|
|
|
98
|
|
|
@pytest.mark.parametrize('mask, expected_idx, expected_element', [ |
|
99
|
|
|
([False, False, False, False, False], 1, 1), |
|
100
|
|
|
([False, True, True, False, False], 3, 3), |
|
101
|
|
|
([False, True, True, True, True], None, None) |
|
102
|
|
|
]) |
|
103
|
|
|
def test_non_masked_elements(mask, expected_idx, expected_element): |
|
104
|
|
|
"""Test with a valid element.""" |
|
105
|
|
|
a = ma.masked_array(np.arange(5), mask=mask) |
|
106
|
|
|
idx, element = _next_non_masked_element(a, 1) |
|
107
|
|
|
assert idx == expected_idx |
|
108
|
|
|
assert element == expected_element |
|
109
|
|
|
|
|
110
|
|
|
|
|
111
|
|
|
@pytest.fixture |
|
112
|
|
|
def thin_point_data(): |
|
113
|
|
|
r"""Provide scattered points for testing.""" |
|
114
|
|
|
xy = np.array([[0.8793620, 0.9005706], [0.5382446, 0.8766988], [0.6361267, 0.1198620], |
|
115
|
|
|
[0.4127191, 0.0270573], [0.1486231, 0.3121822], [0.2607670, 0.4886657], |
|
116
|
|
|
[0.7132257, 0.2827587], [0.4371954, 0.5660840], [0.1318544, 0.6468250], |
|
117
|
|
|
[0.6230519, 0.0682618], [0.5069460, 0.2326285], [0.1324301, 0.5609478], |
|
118
|
|
|
[0.7975495, 0.2109974], [0.7513574, 0.9870045], [0.9305814, 0.0685815], |
|
119
|
|
|
[0.5271641, 0.7276889], [0.8116574, 0.4795037], [0.7017868, 0.5875983], |
|
120
|
|
|
[0.5591604, 0.5579290], [0.1284860, 0.0968003], [0.2857064, 0.3862123]]) |
|
121
|
|
|
return xy |
|
122
|
|
|
|
|
123
|
|
View Code Duplication |
|
|
|
|
|
|
|
124
|
|
|
@pytest.mark.parametrize('radius, truth', |
|
125
|
|
|
[(2.0, np.array([1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
|
126
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 0], dtype=np.bool)), |
|
127
|
|
|
(1.0, np.array([1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
|
128
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 1, 0], dtype=np.bool)), |
|
129
|
|
|
(0.3, np.array([1, 1, 1, 0, 1, 0, 0, 1, 1, 0, 0, |
|
130
|
|
|
0, 0, 0, 0, 0, 1, 0, 0, 0, 0], dtype=np.bool)), |
|
131
|
|
|
(0.1, np.array([1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, |
|
132
|
|
|
0, 1, 1, 1, 1, 1, 1, 1, 1, 1], dtype=np.bool)) |
|
133
|
|
|
]) |
|
134
|
|
|
def test_reduce_point_density(thin_point_data, radius, truth): |
|
135
|
|
|
r"""Test that reduce_point_density works.""" |
|
136
|
|
|
assert_array_equal(reduce_point_density(thin_point_data, radius=radius), truth) |
|
137
|
|
|
|
|
138
|
|
View Code Duplication |
|
|
|
|
|
|
|
139
|
|
|
@pytest.mark.parametrize('radius, truth', |
|
140
|
|
|
[(2.0, np.array([0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
|
141
|
|
|
0, 0, 0, 0, 0, 0, 0, 0, 0, 1], dtype=np.bool)), |
|
142
|
|
|
(0.7, np.array([1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, |
|
143
|
|
|
0, 0, 0, 1, 0, 0, 0, 0, 0, 1], dtype=np.bool)), |
|
144
|
|
|
(0.3, np.array([1, 1, 0, 1, 0, 0, 1, 0, 1, 0, 0, |
|
145
|
|
|
0, 0, 0, 1, 0, 0, 0, 1, 0, 1], dtype=np.bool)), |
|
146
|
|
|
(0.1, np.array([1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, |
|
147
|
|
|
0, 1, 1, 1, 1, 1, 1, 1, 1, 1], dtype=np.bool)) |
|
148
|
|
|
]) |
|
149
|
|
|
def test_reduce_point_density_priority(thin_point_data, radius, truth): |
|
150
|
|
|
r"""Test that reduce_point_density works properly with priority.""" |
|
151
|
|
|
key = np.array([8, 6, 2, 8, 6, 4, 4, 8, 8, 6, 3, 4, 3, 0, 7, 4, 3, 2, 3, 3, 9]) |
|
152
|
|
|
assert_array_equal(reduce_point_density(thin_point_data, radius, key), truth) |
|
153
|
|
|
|
|
154
|
|
|
|
|
155
|
|
|
def test_reduce_point_density_1d(): |
|
156
|
|
|
r"""Test that reduce_point_density works with 1D points.""" |
|
157
|
|
|
x = np.array([1, 3, 4, 8, 9, 10]) |
|
158
|
|
|
assert_array_equal(reduce_point_density(x, 2.5), |
|
159
|
|
|
np.array([1, 0, 1, 1, 0, 0], dtype=np.bool)) |
|
160
|
|
|
|
|
161
|
|
|
|
|
162
|
|
|
def test_delete_masked_points(): |
|
163
|
|
|
"""Test deleting masked points.""" |
|
164
|
|
|
a = ma.masked_array(np.arange(5), mask=[False, True, False, False, False]) |
|
165
|
|
|
b = ma.masked_array(np.arange(5), mask=[False, False, False, True, False]) |
|
166
|
|
|
expected = np.array([0, 2, 4]) |
|
167
|
|
|
a, b = delete_masked_points(a, b) |
|
168
|
|
|
assert_array_equal(a, expected) |
|
169
|
|
|
assert_array_equal(b, expected) |
|
170
|
|
|
|
|
171
|
|
|
|
|
172
|
|
|
def test_log_interp(): |
|
173
|
|
|
"""Test interpolating with log x-scale.""" |
|
174
|
|
|
x_log = np.array([1e3, 1e4, 1e5, 1e6]) |
|
175
|
|
|
y_log = np.log(x_log) * 2 + 3 |
|
176
|
|
|
x_interp = np.array([5e3, 5e4, 5e5]) |
|
177
|
|
|
y_interp_truth = np.array([20.0343863828, 24.6395565688, 29.2447267548]) |
|
178
|
|
|
y_interp = log_interp(x_interp, x_log, y_log) |
|
179
|
|
|
assert_array_almost_equal(y_interp, y_interp_truth, 7) |
|
180
|
|
|
|
|
181
|
|
|
|
|
182
|
|
|
def test_log_interp_units(): |
|
183
|
|
|
"""Test interpolating with log x-scale with units.""" |
|
184
|
|
|
x_log = np.array([1e3, 1e4, 1e5, 1e6]) * units.hPa |
|
185
|
|
|
y_log = (np.log(x_log.m) * 2 + 3) * units.degC |
|
186
|
|
|
x_interp = np.array([5e3, 5e4, 5e5]) * units.hPa |
|
187
|
|
|
y_interp_truth = np.array([20.0343863828, 24.6395565688, 29.2447267548]) * units.degC |
|
188
|
|
|
y_interp = log_interp(x_interp, x_log, y_log) |
|
189
|
|
|
assert_array_almost_equal(y_interp, y_interp_truth, 7) |
|
190
|
|
|
|
Unidata /
MetPy
Loading history...