def test_columns(self):
df = read_occurrence_dataframe(self.f)
self.assertCountEqual(df.columns, [
"buenosaires_usuarios",
"buenosaires_ocurrencias",
"cant_palabra",
"cant_usuarios",
"cant_provincias",
])
def setUp(self):
contents = """
palabra,a_ocurrencias,a_usuarios,b_ocurrencias,b_usuarios
no1,4,1,0,0
no2,7,1,0,0
no3,100,1,20,1
ok1,5,3,0,0
ok2,3,1,3,2
ok3,10,2,20,3
"""
self.f = StringIO(contents)
self.df = read_occurrence_dataframe(self.f, filter_words=(5, 3))
def generate_lists(input_path="output/provinces_words.csv",
output_path="output/listados/"):
"""
Generate ordered lists for each metric words and users
Params:
-------
input_path: string (default="output/provinces_words.csv")
Path to word-provinces matrix
"""
print("Loading words from {}".format(input_path))
df = read_occurrence_dataframe(input_path, filter_words=True)
add_info(df)
df.to_csv(os.path.join(output_path, "listado_completo.csv"))
save_lists(df, output_path)
def test_no_provinces(self):
df = read_occurrence_dataframe(self.f)
self.assertEqual(df.loc["uno"].cant_provincias, 1)
self.assertEqual(df.loc["dos"].cant_provincias, 1)
self.assertEqual(df.loc["tres"].cant_provincias, 2)
def test_no_ocurrences(self):
df = read_occurrence_dataframe(self.f)
self.assertEqual(df.loc["uno"].cant_palabra, 1)
self.assertEqual(df.loc["dos"].cant_palabra, 1)
self.assertEqual(df.loc["tres"].cant_palabra, 4)
def test_no_users(self):
df = read_occurrence_dataframe(self.f)
self.assertEqual(df.loc["uno"].cant_usuarios, 1)
self.assertEqual(df.loc["dos"].cant_usuarios, 1)
self.assertEqual(df.loc["tres"].cant_usuarios, 3)
def test_dataframe_has_three_rows(self):
"""It should remove empty word!"""
df = read_occurrence_dataframe(self.f)
self.assertEqual(len(df), 3)
def test_dataframe_has_no_rows(self):
df = read_occurrence_dataframe(self.f)
self.assertEqual(len(df), 0)