def test_categorical_set_categories():
cat = pd.Categorical(['a', 'a', 'b', 'c', 'a'], categories=['a', 'b', 'c'])
psr = pd.Series(cat)
sr = Series.from_categorical(cat)
# adding category
expect = psr.cat.set_categories(['a', 'b', 'c', 'd'])
got = sr.cat.set_categories(['a', 'b', 'c', 'd'])
assert_eq(expect, got)
# removing category
expect = psr.cat.set_categories(['a', 'b'])
got = sr.cat.set_categories(['a', 'b'])
assert_eq(expect, got)
def test_categorical_set_categories():
cat = pd.Categorical(["a", "a", "b", "c", "a"], categories=["a", "b", "c"])
psr = pd.Series(cat)
sr = Series.from_categorical(cat)
# adding category
expect = psr.cat.set_categories(["a", "b", "c", "d"])
got = sr.cat.set_categories(["a", "b", "c", "d"])
assert_eq(expect, got)
# removing category
expect = psr.cat.set_categories(["a", "b"])
got = sr.cat.set_categories(["a", "b"])
assert_eq(expect, got)
def _make_dictionary_series(self):
"""Make a dictionary-encoded series from this node
"""
assert self.is_dictionary
# create dictionary-encoded column
dict_meta = self.field_schema['dictionary']
dictid = dict_meta['id'] # start from 1
if dict_meta['indexType']['name'] != 'int':
msg = 'non integer type index for dictionary'
raise MetadataParsingError(msg)
ordered = dict_meta['isOrdered']
# find dictionary
for dictionary in self.schema['dictionaries']:
if dictionary['id'] == dictid:
break
categories = dictionary['data']['columns'][0]['DATA']
# make dummy categorical
cat = pd.Categorical([], categories=categories, ordered=ordered)
# make the series
return Series.from_categorical(cat, codes=self.data)
def test_categorical_unique_count(nelem):
from string import ascii_letters, digits
# create categorical series
np.random.seed(12)
pd_cat = pd.Categorical(
pd.Series(np.random.choice(list(ascii_letters + digits), nelem),
dtype='category'))
# gdf
gdf = DataFrame()
gdf['a'] = Series.from_categorical(pd_cat)
gdf_unique_count = gdf['a'].unique_count()
# pandas
pdf = pd.DataFrame()
pdf['a'] = pd_cat
pdf_unique = pdf['a'].unique()
# verify
assert gdf_unique_count == len(pdf_unique)
def test_categorical_unique(num_elements):
from string import ascii_letters, digits
# create categorical series
np.random.seed(12)
pd_cat = pd.Categorical(
pd.Series(np.random.choice(list(ascii_letters + digits), num_elements),
dtype='category'))
# gdf
gdf = DataFrame()
gdf['a'] = Series.from_categorical(pd_cat)
gdf_unique_sorted = np.sort(gdf['a'].unique())
# pandas
pdf = pd.DataFrame()
pdf['a'] = pd_cat
pdf_unique_sorted = np.sort(pdf['a'].unique())
# verify
np.testing.assert_array_equal(pdf_unique_sorted, gdf_unique_sorted)
def test_categorical_value_counts(num_elements):
from string import ascii_letters, digits
# create categorical series
np.random.seed(12)
pd_cat = pd.Categorical(
pd.Series(np.random.choice(list(ascii_letters + digits), num_elements),
dtype='category'))
# gdf
gdf = DataFrame()
gdf['a'] = Series.from_categorical(pd_cat)
gdf_value_counts = gdf['a'].value_counts()
# pandas
pdf = pd.DataFrame()
pdf['a'] = pd_cat
pdf_value_counts = pdf['a'].value_counts()
# verify
pandas_dict = pdf_value_counts.to_dict()
gdf_dict = gdf_value_counts.to_pandas().to_dict()
assert pandas_dict == gdf_dict
