def test_similarity_tree():
# Test that rules are well splitted
rules = [
("a <= 2 and b > 45 and c <= 3 and a > 4", (1, 1, 0)),
("a <= 2 and b > 45 and c <= 3 and a > 4", (1, 1, 0)),
("a > 2 and b > 45", (0.5, 0.3, 0)),
("a > 2 and b > 40", (0.5, 0.2, 0)),
("a <= 2 and b <= 45", (1, 1, 0)),
("a > 2 and c <= 3", (1, 1, 0)),
("b > 45", (1, 1, 0)),
]
sk = SkopeRules(max_depth_duplication=2)
rulesets = sk._find_similar_rulesets(rules)
# Assert some couples of rules are in the same bag
idx_bags_rules = []
for idx_rule, r in enumerate(rules):
idx_bags_for_rule = []
for idx_bag, bag in enumerate(rulesets):
if r in bag:
idx_bags_for_rule.append(idx_bag)
idx_bags_rules.append(idx_bags_for_rule)
assert_equal(idx_bags_rules[0], idx_bags_rules[1])
assert_not_equal(idx_bags_rules[0], idx_bags_rules[2])
# Assert the best rules are kept
final_rules = sk.deduplicate(rules)
assert_in(rules[0], final_rules)
assert_in(rules[2], final_rules)
assert_not_in(rules[3], final_rules)
def test_boundaries():
# ensure min_samples is inclusive of core point
core, _ = dbscan([[0], [1]], eps=2, min_samples=2)
assert_in(0, core)
# ensure eps is inclusive of circumference
core, _ = dbscan([[0], [1], [1]], eps=1, min_samples=2)
assert_in(0, core)
core, _ = dbscan([[0], [1], [1]], eps=.99, min_samples=2)
assert_not_in(0, core)
def test_scorefunc_multilabel():
# Test whether k-best and percentiles works with multilabels with chi2.
X = np.array([[10000, 9999, 0], [100, 9999, 0], [1000, 99, 0]])
y = [[1, 1], [0, 1], [1, 0]]
Xt = SelectKBest(chi2, k=2).fit_transform(X, y)
assert_equal(Xt.shape, (3, 2))
assert_not_in(0, Xt)
Xt = SelectPercentile(chi2, percentile=67).fit_transform(X, y)
assert_equal(Xt.shape, (3, 2))
assert_not_in(0, Xt)
def test_scorefunc_multilabel():
# Test whether k-best and percentiles works with multilabels with chi2.
X = np.array([[10000, 9999, 0], [100, 9999, 0], [1000, 99, 0]])
y = [[1, 1], [0, 1], [1, 0]]
Xt = SelectKBest(chi2, k=2).fit_transform(X, y)
assert_equal(Xt.shape, (3, 2))
assert_not_in(0, Xt)
Xt = SelectPercentile(chi2, percentile=67).fit_transform(X, y)
assert_equal(Xt.shape, (3, 2))
assert_not_in(0, Xt)
def test_tied_pvalues():
# Test whether k-best and percentiles work with tied pvalues from chi2.
# chi2 will return the same p-values for the following features, but it
# will return different scores.
X0 = np.array([[10000, 9999, 9998], [1, 1, 1]])
y = [0, 1]
for perm in itertools.permutations((0, 1, 2)):
X = X0[:, perm]
Xt = SelectKBest(chi2, k=2).fit_transform(X, y)
assert_equal(Xt.shape, (2, 2))
assert_not_in(9998, Xt)
Xt = SelectPercentile(chi2, percentile=67).fit_transform(X, y)
assert_equal(Xt.shape, (2, 2))
assert_not_in(9998, Xt)
def test_tied_pvalues():
"""Test whether k-best and percentiles work with tied pvalues from chi2."""
# chi2 will return the same p-values for the following features, but it
# will return different scores.
X0 = np.array([[10000, 9999, 9998], [1, 1, 1]])
y = [0, 1]
for perm in itertools.permutations((0, 1, 2)):
X = X0[:, perm]
Xt = SelectKBest(chi2, k=2).fit_transform(X, y)
assert_equal(Xt.shape, (2, 2))
assert_not_in(9998, Xt)
Xt = SelectPercentile(chi2, percentile=67).fit_transform(X, y)
assert_equal(Xt.shape, (2, 2))
assert_not_in(9998, Xt)
def test_fetch_one_column():
_urllib2_ref = datasets.mldata.urllib2
try:
dataname = 'onecol'
# create fake data set in cache
x = sp.arange(6).reshape(2, 3)
datasets.mldata.urllib2 = mock_urllib2({dataname: {'x': x}})
dset = fetch_mldata(dataname, data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "data"]:
assert_in(n, dset)
assert_not_in("target", dset)
assert_equal(dset.data.shape, (2, 3))
assert_array_equal(dset.data, x)
# transposing the data array
dset = fetch_mldata(dataname, transpose_data=False, data_home=tmpdir)
assert_equal(dset.data.shape, (3, 2))
finally:
datasets.mldata.urllib2 = _urllib2_ref
def test_fetch_one_column():
_urlopen_ref = datasets.mldata.urlopen
try:
dataname = 'onecol'
# create fake data set in cache
x = sp.arange(6).reshape(2, 3)
datasets.mldata.urlopen = mock_mldata_urlopen({dataname: {'x': x}})
dset = fetch_mldata(dataname, data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "data"]:
assert_in(n, dset)
assert_not_in("target", dset)
assert_equal(dset.data.shape, (2, 3))
assert_array_equal(dset.data, x)
# transposing the data array
dset = fetch_mldata(dataname, transpose_data=False, data_home=tmpdir)
assert_equal(dset.data.shape, (3, 2))
finally:
datasets.mldata.urlopen = _urlopen_ref
def test_fetch_multiple_column():
_urllib2_ref = datasets.mldata.urllib2
try:
# create fake data set in cache
x = sp.arange(6).reshape(2, 3)
y = sp.array([1, -1])
z = sp.arange(12).reshape(4, 3)
# by default
dataname = 'threecol-default'
datasets.mldata.urllib2 = mock_urllib2({
dataname: ({
'label': y,
'data': x,
'z': z
}, ['z', 'data', 'label'])
})
dset = fetch_mldata(dataname, data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "z"]:
assert_in(n, dset)
assert_not_in("x", dset)
assert_not_in("y", dset)
assert_array_equal(dset.data, x)
assert_array_equal(dset.target, y)
assert_array_equal(dset.z, z.T)
# by order
dataname = 'threecol-order'
datasets.mldata.urllib2 = mock_urllib2(
{dataname: ({
'y': y,
'x': x,
'z': z
}, ['y', 'x', 'z'])})
dset = fetch_mldata(dataname, data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "z"]:
assert_in(n, dset)
assert_not_in("x", dset)
assert_not_in("y", dset)
assert_array_equal(dset.data, x)
assert_array_equal(dset.target, y)
assert_array_equal(dset.z, z.T)
# by number
dataname = 'threecol-number'
datasets.mldata.urllib2 = mock_urllib2(
{dataname: ({
'y': y,
'x': x,
'z': z
}, ['z', 'x', 'y'])})
dset = fetch_mldata(dataname,
target_name=2,
data_name=0,
data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "x"]:
assert_in(n, dset)
assert_not_in("y", dset)
assert_not_in("z", dset)
assert_array_equal(dset.data, z)
assert_array_equal(dset.target, y)
# by name
dset = fetch_mldata(dataname,
target_name='y',
data_name='z',
data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "x"]:
assert_in(n, dset)
assert_not_in("y", dset)
assert_not_in("z", dset)
finally:
datasets.mldata.urllib2 = _urllib2_ref
def test_fetch_multiple_column():
_urlopen_ref = datasets.mldata.urlopen
try:
# create fake data set in cache
x = sp.arange(6).reshape(2, 3)
y = sp.array([1, -1])
z = sp.arange(12).reshape(4, 3)
# by default
dataname = 'threecol-default'
datasets.mldata.urlopen = mock_mldata_urlopen({
dataname: (
{
'label': y,
'data': x,
'z': z,
},
['z', 'data', 'label'],
),
})
dset = fetch_mldata(dataname, data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "z"]:
assert_in(n, dset)
assert_not_in("x", dset)
assert_not_in("y", dset)
assert_array_equal(dset.data, x)
assert_array_equal(dset.target, y)
assert_array_equal(dset.z, z.T)
# by order
dataname = 'threecol-order'
datasets.mldata.urlopen = mock_mldata_urlopen({
dataname: ({'y': y, 'x': x, 'z': z},
['y', 'x', 'z']),
})
dset = fetch_mldata(dataname, data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "z"]:
assert_in(n, dset)
assert_not_in("x", dset)
assert_not_in("y", dset)
assert_array_equal(dset.data, x)
assert_array_equal(dset.target, y)
assert_array_equal(dset.z, z.T)
# by number
dataname = 'threecol-number'
datasets.mldata.urlopen = mock_mldata_urlopen({
dataname: ({'y': y, 'x': x, 'z': z},
['z', 'x', 'y']),
})
dset = fetch_mldata(dataname, target_name=2, data_name=0,
data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "x"]:
assert_in(n, dset)
assert_not_in("y", dset)
assert_not_in("z", dset)
assert_array_equal(dset.data, z)
assert_array_equal(dset.target, y)
# by name
dset = fetch_mldata(dataname, target_name='y', data_name='z',
data_home=tmpdir)
for n in ["COL_NAMES", "DESCR", "target", "data", "x"]:
assert_in(n, dset)
assert_not_in("y", dset)
assert_not_in("z", dset)
finally:
datasets.mldata.urlopen = _urlopen_ref
