def test_match_distribution(self):
"""Some really basic testing for match_distribution
"""
ds = datasets['uni2medium'] # large to get stable stats
data = ds.samples[:, ds.a.bogus_features[0]]
# choose bogus feature, which
# should have close to normal distribution
# Lets test ad-hoc rv_semifrozen
floc = rv_semifrozen(scipy.stats.norm, loc=0).fit(data)
self.failUnless(floc[0] == 0)
fscale = rv_semifrozen(scipy.stats.norm, scale=1.0).fit(data)
self.failUnless(fscale[1] == 1)
flocscale = rv_semifrozen(scipy.stats.norm, loc=0, scale=1.0).fit(data)
self.failUnless(flocscale[1] == 1 and flocscale[0] == 0)
full = scipy.stats.norm.fit(data)
for res in [floc, fscale, flocscale, full]:
self.failUnless(len(res) == 2)
data_mean = np.mean(data)
for loc in [None, data_mean]:
for test in ['p-roc', 'kstest']:
# some really basic testing
matched = match_distribution(
data=data,
distributions = ['scipy',
('norm',
{'name': 'norm_fixed',
'loc': 0.2,
'scale': 0.3})],
test=test, loc=loc, p=0.05)
# at least norm should be in there
names = [m[2] for m in matched]
if test == 'p-roc':
if cfg.getboolean('tests', 'labile', default='yes'):
# we can guarantee that only for norm_fixed
self.failUnless('norm' in names)
self.failUnless('norm_fixed' in names)
inorm = names.index('norm_fixed')
# and it should be at least in the first
# 30 best matching ;-)
self.failUnless(inorm <= 30)
# Test plotting only once
if loc is None and externals.exists("pylab plottable"):
import pylab as pl
from mvpa.clfs.stats import plot_distribution_matches
fig = pl.figure()
plot_distribution_matches(data, matched, legend=1, nbest=5)
#pl.show()
pl.close(fig)
def test_match_distribution(self):
"""Some really basic testing for match_distribution
"""
from mvpa.clfs.stats import match_distribution, rv_semifrozen
ds = datasets['uni2medium'] # large to get stable stats
data = ds.samples[:, ds.a.bogus_features[0]]
# choose bogus feature, which
# should have close to normal distribution
# Lets test ad-hoc rv_semifrozen
floc = rv_semifrozen(scipy.stats.norm, loc=0).fit(data)
self.failUnless(floc[0] == 0)
fscale = rv_semifrozen(scipy.stats.norm, scale=1.0).fit(data)
self.failUnless(fscale[1] == 1)
flocscale = rv_semifrozen(scipy.stats.norm, loc=0, scale=1.0).fit(data)
self.failUnless(flocscale[1] == 1 and flocscale[0] == 0)
full = scipy.stats.norm.fit(data)
for res in [floc, fscale, flocscale, full]:
self.failUnless(len(res) == 2)
data_mean = np.mean(data)
for loc in [None, data_mean]:
for test in ['p-roc', 'kstest']:
# some really basic testing
matched = match_distribution(
data=data,
distributions = ['scipy',
('norm',
{'name': 'norm_fixed',
'loc': 0.2,
'scale': 0.3})],
test=test, loc=loc, p=0.05)
# at least norm should be in there
names = [m[2] for m in matched]
if test == 'p-roc':
if cfg.getboolean('tests', 'labile', default='yes'):
# we can guarantee that only for norm_fixed
self.failUnless('norm' in names)
self.failUnless('norm_fixed' in names)
inorm = names.index('norm_fixed')
# and it should be at least in the first
# 30 best matching ;-)
self.failUnless(inorm <= 30)
def test_match_distribution(self):
"""Some really basic testing for match_distribution
"""
from mvpa.clfs.stats import match_distribution, rv_semifrozen
ds = datasets["uni2medium"] # large to get stable stats
data = ds.samples[:, ds.a.bogus_features[0]]
# choose bogus feature, which
# should have close to normal distribution
# Lets test ad-hoc rv_semifrozen
floc = rv_semifrozen(scipy.stats.norm, loc=0).fit(data)
self.failUnless(floc[0] == 0)
fscale = rv_semifrozen(scipy.stats.norm, scale=1.0).fit(data)
self.failUnless(fscale[1] == 1)
flocscale = rv_semifrozen(scipy.stats.norm, loc=0, scale=1.0).fit(data)
self.failUnless(flocscale[1] == 1 and flocscale[0] == 0)
full = scipy.stats.norm.fit(data)
for res in [floc, fscale, flocscale, full]:
self.failUnless(len(res) == 2)
data_mean = np.mean(data)
for loc in [None, data_mean]:
for test in ["p-roc", "kstest"]:
# some really basic testing
matched = match_distribution(
data=data,
distributions=["scipy", ("norm", {"name": "norm_fixed", "loc": 0.2, "scale": 0.3})],
test=test,
loc=loc,
p=0.05,
)
# at least norm should be in there
names = [m[2] for m in matched]
if test == "p-roc":
if cfg.getboolean("tests", "labile", default="yes"):
# we can guarantee that only for norm_fixed
self.failUnless("norm" in names)
self.failUnless("norm_fixed" in names)
inorm = names.index("norm_fixed")
# and it should be at least in the first
# 30 best matching ;-)
self.failUnless(inorm <= 30)
# Prepare few distributions to test
#kwargs = {'permutations':10, 'tail':'any'}
nulldist_sweep = [ MCNullDist(permutations=30, tail='any'),
MCNullDist(permutations=30, tail='right')]
if externals.exists('scipy'):
from mvpa.support.stats import scipy
from scipy.stats import f_oneway
from mvpa.clfs.stats import rv_semifrozen
nulldist_sweep += [ MCNullDist(scipy.stats.norm, permutations=30,
tail='any'),
MCNullDist(scipy.stats.norm, permutations=30,
tail='right'),
MCNullDist(rv_semifrozen(scipy.stats.norm, loc=0),
permutations=30, tail='right'),
MCNullDist(scipy.stats.expon, permutations=30,
tail='right'),
FixedNullDist(scipy.stats.norm(0, 10.0), tail='any'),
FixedNullDist(scipy.stats.norm(0, 10.0), tail='right'),
scipy.stats.norm(0, 0.1)
]
class StatsTests(unittest.TestCase):
"""Unittests for various statistics"""
@sweepargs(null=nulldist_sweep[1:])
def test_null_dist_prob(self, null):
"""Testing null dist probability"""
# Prepare few distributions to test
#kwargs = {'permutations':10, 'tail':'any'}
permutator = AttributePermutator('targets', count=30)
nulldist_sweep = [ MCNullDist(permutator, tail='any'),
MCNullDist(permutator, tail='right')]
if externals.exists('scipy'):
from mvpa.support.stats import scipy
from scipy.stats import f_oneway
from mvpa.clfs.stats import rv_semifrozen
nulldist_sweep += [ MCNullDist(permutator, scipy.stats.norm,
tail='any'),
MCNullDist(permutator, scipy.stats.norm,
tail='right'),
MCNullDist(permutator,
rv_semifrozen(scipy.stats.norm, loc=0),
tail='right'),
MCNullDist(permutator, scipy.stats.expon,
tail='right'),
FixedNullDist(scipy.stats.norm(0, 10.0), tail='any'),
FixedNullDist(scipy.stats.norm(0, 10.0), tail='right'),
scipy.stats.norm(0, 0.1)
]
class StatsTests(unittest.TestCase):
"""Unittests for various statistics"""
@sweepargs(null=nulldist_sweep[1:])
def test_null_dist_prob(self, null):
"""Testing null dist probability"""
# Prepare few distributions to test
#kwargs = {'permutations':10, 'tail':'any'}
nulldist_sweep = [
MCNullDist(permutations=30, tail='any'),
MCNullDist(permutations=30, tail='right')
]
if externals.exists('scipy'):
from mvpa.support.stats import scipy
from scipy.stats import f_oneway
from mvpa.clfs.stats import rv_semifrozen
nulldist_sweep += [
MCNullDist(scipy.stats.norm, permutations=30, tail='any'),
MCNullDist(scipy.stats.norm, permutations=30, tail='right'),
MCNullDist(rv_semifrozen(scipy.stats.norm, loc=0),
permutations=30,
tail='right'),
MCNullDist(scipy.stats.expon, permutations=30, tail='right'),
FixedNullDist(scipy.stats.norm(0, 10.0), tail='any'),
FixedNullDist(scipy.stats.norm(0, 10.0), tail='right'),
scipy.stats.norm(0, 0.1)
]
class StatsTests(unittest.TestCase):
"""Unittests for various statistics"""
@sweepargs(null=nulldist_sweep[1:])
def test_null_dist_prob(self, null):
"""Testing null dist probability"""
if not isinstance(null, NullDist):