import numpy as np
import pickle
from .testutil import datadir
from brainstat.stats._multiple_comparisons import stat_threshold
[docs]def dummy_test(infile, expfile):
# load input test data
ifile = open(infile, "br")
idic = pickle.load(ifile)
ifile.close()
# run stat_threshold
A, B, C, D, E, F = stat_threshold(
idic["search_volume"],
idic["num_voxels"],
idic["fwhm"],
idic["df"],
idic["p_val_peak"],
idic["cluster_threshold"],
idic["p_val_extent"],
idic["nconj"],
idic["nvar"],
None,
None,
idic["nprint"],
)
outdic = {
"peak_threshold": A,
"extent_threshold": B,
"peak_threshold_1": C,
"extent_threshold_1": D,
"t": E,
"rho": F,
}
# load expected outout data
efile = open(expfile, "br")
expdic = pickle.load(efile)
efile.close()
testout = []
for key in outdic.keys():
comp = np.allclose(outdic[key], expdic[key], rtol=1e-05, equal_nan=True)
testout.append(comp)
assert all(flag == True for (flag) in testout)
# parameters in *pck is equal to default params, if not specified in tests
[docs]def test_01():
# search_volume = 5.59, df = 5, nconj = 0.5, nprint = 0
infile = datadir("thresh_01_IN.pkl")
expfile = datadir("thresh_01_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_02():
# search volume is a list, search volume = [3.2, 8.82, 7.71],
# df = 5, nconj = 0.5, nprint=0
infile = datadir("thresh_02_IN.pkl")
expfile = datadir("thresh_02_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_03():
# search volume is a 1D numpy array, search volume = np.array([0.36, 6.22 , 2.13]),
# df = 5, nconj = 0.5, nprint=0
infile = datadir("thresh_03_IN.pkl")
expfile = datadir("thresh_03_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_04():
# search_volume is a 2D numpy array,
# search_volume = array([[ 0.46, 3.00], [13.00, 87.46]]),
# df = 5, nconj = 0.5, nprint=0
infile = datadir("thresh_04_IN.pkl")
expfile = datadir("thresh_04_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_05():
# search_volume is a float, search_volume = 1.22,
# num_voxels = 560, df = 5, nconj = 0.5, nprint = 0
infile = datadir("thresh_05_IN.pkl")
expfile = datadir("thresh_05_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_06():
# search_volume is a 2D numpy array,
# search volume = array([[3955, 760], [ 770, 4042]]),
# num_voxels is a list of integers, num_voxels = [3955, 760, 8058],
# df = 5, nconj = 0.5, nprint = 0
infile = datadir("thresh_06_IN.pkl")
expfile = datadir("thresh_06_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_07():
# search_volume is a 2D numpy array,
# search_volume = array([[ 49, 47], [ 57, 136]]),
# num_voxels is a 2D array of shape (23, 1), including floats,
# fwhm = 5.34, df = 35, nconj = 0.5, nprint=0
infile = datadir("thresh_07_IN.pkl")
expfile = datadir("thresh_07_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_08():
# search_volume is a 2D numpy array, array([[ 76, 71], [ 81, 163]]),
# num_voxels is a 2D array of shape (75, 1), dtype('float64'),
# fwhm = 3.57, df = inf, nconj = 0.5, nprint=0
infile = datadir("thresh_08_IN.pkl")
expfile = datadir("thresh_08_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_09():
# search_volume is a float, search_volume = 70.66,
# num_voxels is a 1D array of shape (5,), dtype('int64'),
# fwhm = 6.95, df = 5, p_val_peak = 0.71, nconj = 0.5, nprint=0
infile = datadir("thresh_09_IN.pkl")
expfile = datadir("thresh_09_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_10():
# search_volume is a float, search_volume = 72.70,
# num_voxels is a 1D array of shape (5,), dtype('int64'),
# fwhm = 5.21, df = 7,
# p_val_peak is a list, p_val_peak = [0.48, 0.54, 0.24, 0.27],
# nconj = 0.5, nprint=0
infile = datadir("thresh_10_IN.pkl")
expfile = datadir("thresh_10_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_11():
# search_volume is a float, search_volume = 33.26,
# num_voxels is a 1D array of shape (5,), dtype('int64'),
# fwhm = 5.20, df = 7,
# p_val_peak is a 1D array of shape (7,), dtype('float64'),
# nconj = 0.5, nprint=0
infile = datadir("thresh_11_IN.pkl")
expfile = datadir("thresh_11_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_12():
# search_volume is a float, search_volume = 7.0,
# num_voxels = 55, fwhm = 9.50, df = 6,
# p_val_peak is a 1D array of shape (55,), dtype('float64'),
# cluster_threshold = 0.048,
# p_val_extent is a 1D array of shape (55,), dtype('float64'),
# nconj = 54, nprint = 0
infile = datadir("thresh_12_IN.pkl")
expfile = datadir("thresh_12_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_13():
# search_volume is a list, search_volume = [5.7, 8, 9],
# num_voxels = 100, df = 5,
# p_val_peak is a 2D array of shape (2,2), dtype('float64'),
# cluster_threshold is a list, cluster_threshold = [0.001, 0.1, 0, 6]
# p_val_extent = 0.01, nconj = 0.02, nvar = [1, 1], nprint = 0
infile = datadir("thresh_13_IN.pkl")
expfile = datadir("thresh_13_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_14():
# search_volume is a 2D numpy array,
# search_volume = np.array([[4.00e+00, nan, 6.59e+03]]).
# num_voxels = 64984, fwhm = 1,
# df a 2D array of shape (2,2), dtype('int64'),
# p_val_peak is a 1D array of shape (2,), dtype('float64'),
# cluster_threshold = 0.001, p_val_extent = 0.05, nprint = 1
infile = datadir("thresh_14_IN.pkl")
expfile = datadir("thresh_14_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_15():
# df a 2D array of shape (2, 2), dtype('int64'),
# p_val_peak is a 2D array of shape (1, 87), dtype('float64'),
# nprint = 0
infile = datadir("thresh_15_IN.pkl")
expfile = datadir("thresh_15_OUT.pkl")
dummy_test(infile, expfile)
[docs]def test_16():
# df a 2D array of shape (2, 2), dtype('int64'),
# p_val_peak = 0.001, nvar = 3.0 nprint = 0
infile = datadir("thresh_16_IN.pkl")
expfile = datadir("thresh_16_OUT.pkl")
dummy_test(infile, expfile)