Skip to content

Commit

Permalink
fix missing test
Browse files Browse the repository at this point in the history
  • Loading branch information
Benjamin Alan Weaver committed Nov 13, 2015
1 parent 9d71bde commit 4dd6bbb
Showing 1 changed file with 92 additions and 88 deletions.
180 changes: 92 additions & 88 deletions pydl/pydlutils/tests/test_math.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@
# -*- coding: utf-8 -*-
import numpy as np
from astropy.tests.helper import raises
from ..math import computechi2, djs_median
from ..math import computechi2, djs_median, find_contiguous


class TestMath(object):
Expand Down Expand Up @@ -41,90 +41,94 @@ def test_computechi2(self):
#
assert (chi2.var == np.diag(chi2.covar)).all()

def test_djs_median(self):
np.random.seed(424242)
data = 100.0*np.random.random(100)
data2 = 100.0*np.random.random((10, 10))
data3 = 100.0*np.random.random((10, 10, 10))
data_width_5 = np.array([
49.68311576, 74.83671757, 49.68311576, 42.12573137,
32.11576752, 27.22092569, 15.08905903, 15.08905903,
27.22092569, 20.07809411, 41.55978953, 41.55978953,
35.01184343, 35.01184343, 38.7967727 , 38.7967727 ,
38.7967727 , 38.7967727 , 38.7967727 , 28.85780425,
28.85780425, 30.23801035, 30.23801035, 27.0687078 ,
30.23801035, 64.77058052, 29.42407045, 38.51749618,
62.48793217, 38.51749618, 29.42407045, 38.51749618,
18.46919414, 18.46919414, 42.72395431, 54.11584807,
54.11584807, 75.67593452, 75.67593452, 72.4380356 ,
61.68761955, 61.68761955, 45.36796557, 45.36796557,
61.68761955, 76.57621138, 76.57621138, 76.57621138,
23.28589488, 23.28589488, 13.82755909, 12.10607597,
13.82755909, 27.51891089, 44.21266068, 44.21266068,
44.21266068, 47.18083025, 47.18083025, 47.18083025,
47.18083025, 47.18083025, 35.50809933, 35.50809933,
25.52222293, 25.52222293, 67.8568479 , 88.54983822,
67.8568479 , 93.40053148, 93.40053148, 64.12514945,
47.82074715, 47.82074715, 47.82074715, 34.82234113,
34.82234113, 52.91092248, 78.51075522, 92.16442338,
92.16442338, 92.16442338, 72.07989558, 72.07989558,
68.72579501, 72.07989558, 72.07989558, 70.43134908,
34.55273356, 62.09010468, 62.09010468, 70.43134908,
68.89705132, 68.89705132, 68.89705132, 66.30426084,
55.92748086, 55.92748086, 55.92748086, 65.11387444])
data_width_5_reflect = np.array([
49.68311576, 49.68311576, 49.68311576, 42.12573137,
32.11576752, 27.22092569, 15.08905903, 15.08905903,
27.22092569, 20.07809411, 41.55978953, 41.55978953,
35.01184343, 35.01184343, 38.7967727 , 38.7967727 ,
38.7967727 , 38.7967727 , 38.7967727 , 28.85780425,
28.85780425, 30.23801035, 30.23801035, 27.0687078 ,
30.23801035, 64.77058052, 29.42407045, 38.51749618,
62.48793217, 38.51749618, 29.42407045, 38.51749618,
18.46919414, 18.46919414, 42.72395431, 54.11584807,
54.11584807, 75.67593452, 75.67593452, 72.4380356 ,
61.68761955, 61.68761955, 45.36796557, 45.36796557,
61.68761955, 76.57621138, 76.57621138, 76.57621138,
23.28589488, 23.28589488, 13.82755909, 12.10607597,
13.82755909, 27.51891089, 44.21266068, 44.21266068,
44.21266068, 47.18083025, 47.18083025, 47.18083025,
47.18083025, 47.18083025, 35.50809933, 35.50809933,
25.52222293, 25.52222293, 67.8568479 , 88.54983822,
67.8568479 , 93.40053148, 93.40053148, 64.12514945,
47.82074715, 47.82074715, 47.82074715, 34.82234113,
34.82234113, 52.91092248, 78.51075522, 92.16442338,
92.16442338, 92.16442338, 72.07989558, 72.07989558,
68.72579501, 72.07989558, 72.07989558, 70.43134908,
34.55273356, 62.09010468, 62.09010468, 70.43134908,
68.89705132, 68.89705132, 68.89705132, 55.92748086,
65.11387444, 55.92748086, 55.92748086, 55.92748086])
#
# Degenerate cases that fall back on numpy.median().
#
assert np.allclose(np.median(data), djs_median(data))
assert np.allclose(np.median(data2, axis=0),
djs_median(data2, dimension=0))
assert np.allclose(np.median(data2, axis=1),
djs_median(data2, dimension=1))
#
# Test widths.
#
assert np.allclose(data, djs_median(data, width=1))
assert np.allclose(data_width_5, djs_median(data, width=5))
# assert np.allclose(data_width_5_reflect,
# djs_median(data, width=5, boundary='reflect'))
#
# Exceptions
#
with raises(ValueError):
foo = djs_median(data, width=5, boundary='nearest')
with raises(ValueError):
foo = djs_median(data, width=5, boundary='wrap')
with raises(ValueError):
foo = djs_median(data, width=5, boundary='foobar')
with raises(ValueError):
foo = djs_median(data2, width=5, dimension=1)
with raises(ValueError):
foo = djs_median(data3, width=5)
with raises(ValueError):
foo = djs_median(data3, width=5, boundary='reflect')
def test_djs_median(self):
np.random.seed(424242)
data = 100.0*np.random.random(100)
data2 = 100.0*np.random.random((10, 10))
data3 = 100.0*np.random.random((10, 10, 10))
data_width_5 = np.array([
49.68311576, 74.83671757, 49.68311576, 42.12573137,
32.11576752, 27.22092569, 15.08905903, 15.08905903,
27.22092569, 20.07809411, 41.55978953, 41.55978953,
35.01184343, 35.01184343, 38.7967727 , 38.7967727 ,
38.7967727 , 38.7967727 , 38.7967727 , 28.85780425,
28.85780425, 30.23801035, 30.23801035, 27.0687078 ,
30.23801035, 64.77058052, 29.42407045, 38.51749618,
62.48793217, 38.51749618, 29.42407045, 38.51749618,
18.46919414, 18.46919414, 42.72395431, 54.11584807,
54.11584807, 75.67593452, 75.67593452, 72.4380356 ,
61.68761955, 61.68761955, 45.36796557, 45.36796557,
61.68761955, 76.57621138, 76.57621138, 76.57621138,
23.28589488, 23.28589488, 13.82755909, 12.10607597,
13.82755909, 27.51891089, 44.21266068, 44.21266068,
44.21266068, 47.18083025, 47.18083025, 47.18083025,
47.18083025, 47.18083025, 35.50809933, 35.50809933,
25.52222293, 25.52222293, 67.8568479 , 88.54983822,
67.8568479 , 93.40053148, 93.40053148, 64.12514945,
47.82074715, 47.82074715, 47.82074715, 34.82234113,
34.82234113, 52.91092248, 78.51075522, 92.16442338,
92.16442338, 92.16442338, 72.07989558, 72.07989558,
68.72579501, 72.07989558, 72.07989558, 70.43134908,
34.55273356, 62.09010468, 62.09010468, 70.43134908,
68.89705132, 68.89705132, 68.89705132, 66.30426084,
55.92748086, 55.92748086, 55.92748086, 65.11387444])
data_width_5_reflect = np.array([
49.68311576, 49.68311576, 49.68311576, 42.12573137,
32.11576752, 27.22092569, 15.08905903, 15.08905903,
27.22092569, 20.07809411, 41.55978953, 41.55978953,
35.01184343, 35.01184343, 38.7967727 , 38.7967727 ,
38.7967727 , 38.7967727 , 38.7967727 , 28.85780425,
28.85780425, 30.23801035, 30.23801035, 27.0687078 ,
30.23801035, 64.77058052, 29.42407045, 38.51749618,
62.48793217, 38.51749618, 29.42407045, 38.51749618,
18.46919414, 18.46919414, 42.72395431, 54.11584807,
54.11584807, 75.67593452, 75.67593452, 72.4380356 ,
61.68761955, 61.68761955, 45.36796557, 45.36796557,
61.68761955, 76.57621138, 76.57621138, 76.57621138,
23.28589488, 23.28589488, 13.82755909, 12.10607597,
13.82755909, 27.51891089, 44.21266068, 44.21266068,
44.21266068, 47.18083025, 47.18083025, 47.18083025,
47.18083025, 47.18083025, 35.50809933, 35.50809933,
25.52222293, 25.52222293, 67.8568479 , 88.54983822,
67.8568479 , 93.40053148, 93.40053148, 64.12514945,
47.82074715, 47.82074715, 47.82074715, 34.82234113,
34.82234113, 52.91092248, 78.51075522, 92.16442338,
92.16442338, 92.16442338, 72.07989558, 72.07989558,
68.72579501, 72.07989558, 72.07989558, 70.43134908,
34.55273356, 62.09010468, 62.09010468, 70.43134908,
68.89705132, 68.89705132, 68.89705132, 55.92748086,
65.11387444, 55.92748086, 55.92748086, 55.92748086])
#
# Degenerate cases that fall back on numpy.median().
#
assert np.allclose(np.median(data), djs_median(data))
assert np.allclose(np.median(data2, axis=0),
djs_median(data2, dimension=0))
assert np.allclose(np.median(data2, axis=1),
djs_median(data2, dimension=1))
#
# Test widths.
#
assert np.allclose(data, djs_median(data, width=1))
assert np.allclose(data_width_5, djs_median(data, width=5))
# assert np.allclose(data_width_5_reflect,
# djs_median(data, width=5, boundary='reflect'))
#
# Exceptions
#
with raises(ValueError):
foo = djs_median(data, width=5, boundary='nearest')
with raises(ValueError):
foo = djs_median(data, width=5, boundary='wrap')
with raises(ValueError):
foo = djs_median(data, width=5, boundary='foobar')
with raises(ValueError):
foo = djs_median(data2, width=5, dimension=1)
with raises(ValueError):
foo = djs_median(data3, width=5)
with raises(ValueError):
foo = djs_median(data3, width=5, boundary='reflect')


def test_find_contiguous(self):
assert find_contiguous(np.array([0,1,1,1,0,1,1,0,1])) == [1,2,3]

0 comments on commit 4dd6bbb

Please sign in to comment.