def test_guess_dtypes_mixed_types(smalldf):
dtypes = du.guess_dtypes(smalldf)
assert dtypes[0] == 'continuous'
assert dtypes[1] == 'categorical'
assert dtypes[2] == 'categorical'
assert dtypes[3] == 'continuous'
def test_guess_dtypes_mixed_types(smalldf):
dtypes = du.guess_dtypes(smalldf)
assert dtypes[0] == 'continuous'
assert dtypes[1] == 'categorical'
assert dtypes[2] == 'categorical'
assert dtypes[3] == 'continuous'
def test_guess_dtypes_mixed_types_missing_vals(smalldf):
smalldf.ix[0, 0] = float('NaN')
smalldf.ix[0, 1] = float('NaN')
smalldf.ix[0, 2] = float('NaN')
dtypes = du.guess_dtypes(smalldf)
assert dtypes[0] == 'continuous'
assert dtypes[1] == 'categorical'
assert dtypes[2] == 'categorical'
assert dtypes[3] == 'continuous'
def test_guess_dtypes_mixed_types_missing_vals(smalldf):
smalldf.ix[0, 0] = float('NaN')
smalldf.ix[0, 1] = float('NaN')
smalldf.ix[0, 2] = float('NaN')
dtypes = du.guess_dtypes(smalldf)
assert dtypes[0] == 'continuous'
assert dtypes[1] == 'categorical'
assert dtypes[2] == 'categorical'
assert dtypes[3] == 'continuous'
def test_guess_dtypes_decrease_unique_vals_cutoff():
# large number of unique values
df = pd.DataFrame(np.random.rand(5, 4))
dtypes = du.guess_dtypes(df, n_unique_cutoff=2)
assert all([dt == 'continuous' for dt in dtypes])
def test_guess_dtypes_increase_unique_vals_cutoff():
# large number of unique values
df = pd.DataFrame(np.random.rand(30, 4))
dtypes = du.guess_dtypes(df, n_unique_cutoff=32)
assert all([dt == 'categorical' for dt in dtypes])
def test_guess_dtypes_should_guess_correct_types_continuous_short():
# small number of unique values
df = pd.DataFrame(np.random.rand(5, 4))
dtypes = du.guess_dtypes(df)
assert all([dt == 'categorical' for dt in dtypes])
def test_guess_dtypes_should_guess_correct_types_continuous():
df = pd.DataFrame(np.random.rand(30, 4))
dtypes = du.guess_dtypes(df)
assert all([dt == 'continuous' for dt in dtypes])
def test_guess_dtypes_decrease_unique_vals_cutoff():
# large number of unique values
df = pd.DataFrame(np.random.rand(5, 4))
dtypes = du.guess_dtypes(df, n_unique_cutoff=2)
assert all([dt == 'continuous' for dt in dtypes])
def test_guess_dtypes_increase_unique_vals_cutoff():
# large number of unique values
df = pd.DataFrame(np.random.rand(30, 4))
dtypes = du.guess_dtypes(df, n_unique_cutoff=32)
assert all([dt == 'categorical' for dt in dtypes])
def test_guess_dtypes_should_guess_correct_types_continuous_short():
# small number of unique values
df = pd.DataFrame(np.random.rand(5, 4))
dtypes = du.guess_dtypes(df)
assert all([dt == 'categorical' for dt in dtypes])
def test_guess_dtypes_should_guess_correct_types_continuous():
df = pd.DataFrame(np.random.rand(30, 4))
dtypes = du.guess_dtypes(df)
assert all([dt == 'continuous' for dt in dtypes])
