def test_query_transform_mean():
original = pd.DataFrame(
{
'distance': [20, 10, 30, 40],
'token': ['t1', 't1', 't2', 't2']
},
index=['a', 'b', 'c', 'd'])
transformed = query_transform(PandasTable(original),
{'query': 'distance < mean[distance]'})
expected = pd.DataFrame({
'distance': [20, 10],
'token': ['t1', 't1']
},
index=['a', 'b'])
assert expected.equals(transformed.to_pandas())
def test_histogram_transform_numeric():
original = pd.DataFrame({'distance': [20, 10, 30, 40]},
index=['a', 'b', 'c', 'd'])
transformed = histogram_transform(PandasTable(original), {
'column': 'distance',
'bins': 2
})
expected_side_result = {
'x': [17.5, 32.5],
'y': [2, 2],
'width': [15., 15.]
}
assert original.equals(transformed.to_pandas())
assert np.allclose(expected_side_result['x'], transformed.side_result['x'])
assert np.allclose(expected_side_result['y'], transformed.side_result['y'])
assert np.allclose(expected_side_result['width'],
transformed.side_result['width'])
def test_search_transform_string():
original = pd.DataFrame(
{
'distance': [20, 10, 30, 40],
'token': ['t1', 't1', 't2', 't2']
},
index=['a', 'b', 'c', 'd'])
transformed = search_transform(PandasTable(original), {
'column': 'token',
'searchterm': 't2'
})
expected = pd.DataFrame({
'distance': [30, 40],
'token': ['t2', 't2']
},
index=['c', 'd'])
assert expected.equals(transformed.to_pandas())
def test_histogram_transform_categorical():
original = pd.DataFrame({'token': ['t1', 't1', 't2', 't2']},
index=['a', 'b', 'c', 'd'])
transformed = histogram_transform(PandasTable(original), {
'column': 'token',
'bins': 2
})
expected_side_result = {
'x': ['t1', 't2', 'Other'],
'y': [2, 2, 0],
'width': None
}
assert original.equals(transformed.to_pandas())
assert expected_side_result.keys() == transformed.side_result.keys()
assert expected_side_result['width'] == transformed.side_result['width']
assert set(expected_side_result['x']) == set(
transformed.side_result['x']) # 't1', 't2' may be flipped
assert expected_side_result['y'] == transformed.side_result['y']
def test_sample_transform():
original = pd.DataFrame({'distance': [20, 10, 30]}, index=['a', 'b', 'c'])
transformed = sample_transform(PandasTable(original), {'fraction': 1.0})
expected = pd.DataFrame({'distance': [20, 10, 30]}, index=['a', 'b', 'c'])
assert expected.equals(transformed.to_pandas())
def test_pandas_table_apply_bounds():
original = pd.DataFrame({'distance': [20, 10, 30, 40]},
index=['a', 'b', 'c', 'd'])
expected = pd.DataFrame({'distance': [10, 30]}, index=['b', 'c'])
assert expected.equals(
PandasTable(original).apply_bounds(1, 2).to_pandas())
def test_pandas_table_to_pandas():
original = pd.DataFrame({'distance': [20, 10, 30, 40]},
index=['a', 'b', 'c', 'd'])
assert original.equals(PandasTable(original).to_pandas())
def test_pandas_table_len():
original = pd.DataFrame({'distance': [20, 10, 30, 40]},
index=['a', 'b', 'c', 'd'])
assert 4 == len(PandasTable(original))
def test_sort_transform():
original = pd.DataFrame({'distance': [20, 10, 30]}, index=['a', 'b', 'c'])
transformed = sort_transform(PandasTable(original), {'column': 'distance', 'ascending': False})
expected = pd.DataFrame({'distance': [30, 20, 10]}, index=['c', 'a', 'b'])
assert expected.equals(transformed.to_pandas())