def test_mean_nan(es):
array = pd.Series([5, 5, 5, 5, 5])
mean_func_nans_default = Mean().get_function()
mean_func_nans_false = Mean(skipna=False).get_function()
mean_func_nans_true = Mean(skipna=True).get_function()
assert mean_func_nans_default(array) == 5
assert mean_func_nans_false(array) == 5
assert mean_func_nans_true(array) == 5
array = pd.Series([5, np.nan, np.nan, np.nan, np.nan, 10])
assert mean_func_nans_default(array) == 7.5
assert isnan(mean_func_nans_false(array))
assert mean_func_nans_true(array) == 7.5
array_nans = pd.Series([np.nan, np.nan, np.nan, np.nan])
assert isnan(mean_func_nans_default(array_nans))
assert isnan(mean_func_nans_false(array_nans))
assert isnan(mean_func_nans_true(array_nans))
# test naming
default_feat = ft.Feature(es["log"]["value"],
parent_entity=es["customers"],
primitive=Mean)
assert default_feat.get_name() == "MEAN(log.value)"
ignore_nan_feat = ft.Feature(es["log"]["value"],
parent_entity=es["customers"],
primitive=Mean(skipna=True))
assert ignore_nan_feat.get_name() == "MEAN(log.value)"
include_nan_feat = ft.Feature(es["log"]["value"],
parent_entity=es["customers"],
primitive=Mean(skipna=False))
assert include_nan_feat.get_name() == "MEAN(log.value, skipna=False)"
def test_check_input_types(es):
count = Count(es["sessions"]["id"], es["customers"])
mean = Mean(count, es[u"régions"])
assert mean._check_input_types()
boolean = count > 3
mean = Mean(count, es[u"régions"], where=boolean)
assert mean._check_input_types()
def test_mean_nan():
array = np.array([5, 5, 5, 5, 5])
mean_func_nans_default = Mean().get_function()
mean_func_nans_false = Mean(ignore_nans=False).get_function()
mean_func_nans_true = Mean(ignore_nans=True).get_function()
assert mean_func_nans_default(array) == 5
assert mean_func_nans_false(array) == 5
assert mean_func_nans_true(array) == 5
array = np.array([5, np.nan, np.nan, np.nan, np.nan, 10])
assert isnan(mean_func_nans_default(array))
assert isnan(mean_func_nans_false(array))
assert mean_func_nans_true(array) == 7.5
array_nans = np.array([np.nan, np.nan, np.nan, np.nan])
assert isnan(mean_func_nans_default(array))
assert isnan(mean_func_nans_false(array_nans))
assert isnan(mean_func_nans_true(array_nans))
def test_make_deep_agg_feat_of_dfeat_of_agg_feat(entityset, backend):
"""
The graph looks like this (higher implies parent):
C C = Customers, the entity we're trying to predict on
| S = Sessions, a child of Customers
P S L = Log, a child of both Sessions and Log
\ / P = Products, a parent of Log which is not a descendent of customers
L
We're trying to calculate a DFeat from L to P on an agg_feat of P on L, and
then aggregate it with another agg_feat of C on L.
"""
log_count_feat = Count(entityset['log']['id'],
parent_entity=entityset['products'])
product_purchases_feat = DirectFeature(log_count_feat,
child_entity=entityset['log'])
purchase_popularity = Mean(product_purchases_feat,
parent_entity=entityset['customers'])
pandas_backend = backend([purchase_popularity])
df = pandas_backend.calculate_all_features(instance_ids=[0],
time_last=None)
v = df[purchase_popularity.get_name()][0]
assert (v == 38.0 / 10.0)
def test_make_compare_feat(entityset, backend):
"""
Feature we're creating is:
Number of sessions for each customer where the
number of logs in the session is less than 3
"""
Count.max_stack_depth = 2
log_count_feat = Count(entityset['log']['id'],
parent_entity=entityset['sessions'])
mean_agg_feat = Mean(log_count_feat, parent_entity=entityset['customers'])
mean_feat = DirectFeature(mean_agg_feat,
child_entity=entityset['sessions'])
feat = log_count_feat > mean_feat
pandas_backend = backend([feat])
df = pandas_backend.calculate_all_features(instance_ids=[0, 1, 2],
time_last=None)
name = feat.get_name()
instances = df[name]
v0, v1, v2 = instances[0:3]
assert v0
assert v1
assert not v2
def test_deep_agg_feat_chain(entityset, backend):
"""
Agg feat of agg feat:
region.Mean(customer.Count(Log))
"""
customer_count_feat = Count(entityset['log']['id'],
parent_entity=entityset['customers'])
region_avg_feat = Mean(customer_count_feat,
parent_entity=entityset[u'régions'])
pandas_backend = backend([region_avg_feat])
df = pandas_backend.calculate_all_features(instance_ids=['United States'],
time_last=None)
v = df[region_avg_feat.get_name()][0]
assert (v == 17 / 3.)