def test_lazy_slice(self):
data = Data([1, 2, 3, 4])
assert list(data) == [1, 2, 3, 4]
assert data.index == range(0, 4)
assert list(data.lazy_slice([3, 2, 1])) == [4, 3, 2]
assert data.index == range(0, 4)
def get_data(self):
data = self._parse_json()
x = self._extract_features(data)
y = self._extract_labels(data)
index = x.index
data = Container(
{
'y': Data(y.values, index=index),
'x': Data(x.values, index=index, dtype=float64)
},
index=index)
return data
def test_can_make_data_container_from_dict_of_data(self):
x = [[1, 2], [2, 3], [3, 4]]
y = [0, 1, 0]
data = Container.from_dict(
{
'x': Data(x, index=range(3), shape=[2]),
'y': Data(y, index=range(3), shape=[0]),
},
index=range(3)
)
assert isinstance(data, Container)
assert isinstance(data, Data)
assert isinstance(data['x'], Data)
assert isinstance(data['y'], Data)
assert not isinstance(data['x'].data, Data)
assert np.array_equal(data['x'].as_numpy, np.array(x))
def get_data(self):
x, y = make_classification(**self.kwargs)
index = range(len(x))
x = Data(x, index=index, dtype=float64)
y = Data(y, index=index, dtype=float64)
return Container.from_dict({'x': x, 'y': y}, index=index)
def test_data_shape(self):
data = Data([1, 2, 3, 4])
assert data.shape == tf.TensorShape([])
data = Data([[1, 2], [3, 4]])
assert data.shape == tf.TensorShape([2])
def test_as_numpy(self):
data = Data([1, 2, 3, 4])
assert np.array_equal(data.as_numpy, np.array([1, 2, 3, 4]))
def test_split(self):
data = Data([1, 2, 3, 4])
x, y = data.split([1, 2], [3, 0])
assert list(x) == [2, 3]
assert list(y) == [4, 1]
def test_lazy_slice_works_on_array(self):
data = Data([[1, 2, 3], [2, 3, 4]], index=[0, 1, 2])
assert list(data.lazy_slice([0])) == [[1, 2, 3]]
assert data[1] == [2, 3, 4]