def testUnzip(self):
n1 = numpy.array([[1., 2.], [3., 4.], [5., 6.], [7., 8.]],
dtype=numpy.float32)
t1 = tf.constant(n1)
out = self.Run(functions.unzip(t1, 0, 4, 2))
expected = numpy.array([[1., 2.], [5., 6.]], dtype=numpy.float32)
testing.assert_allclose(expected, out[0], rtol=TOLERANCE)
expected = numpy.array([[3., 4.], [7., 8.]], dtype=numpy.float32)
testing.assert_allclose(expected, out[1], rtol=TOLERANCE)
def testUnzip(self):
n1 = numpy.array([[1., 2.], [3., 4.], [5., 6.], [7., 8.]],
dtype=numpy.float32)
t1 = tf.constant(n1)
out = self.Run(functions.unzip(t1, 0, 4, 2))
expected = numpy.array([[1., 2.], [5., 6.]], dtype=numpy.float32)
testing.assert_allclose(expected, out[0], rtol=TOLERANCE)
expected = numpy.array([[3., 4.], [7., 8.]], dtype=numpy.float32)
testing.assert_allclose(expected, out[1], rtol=TOLERANCE)
def unzip(input_layer, split_dim=0, num_splits=2):
"""Unzips the head Tensor along the split_dim into num_splits Equal chunks.
Examples:
* `[1, 2, 3, 4] -> [1, 3], [2, 4]`
* `[[1, 1], [2, 2], [3, 3], [4, 4]] -> [[1, 1], [3, 3]], [[2, 2], [4, 4]]`
Args:
input_layer: The chainable object, supplied.
split_dim: The dimension to split along. Defaults to batch.
num_splits: The number of splits.
Returns:
A list of PrettyTensors.
Raises:
ValueError: If split_dim is out of range or isn't divided evenly by
num_splits.
"""
shape = input_layer.shape
_check_split_dims(num_splits, split_dim, shape)
splits = functions.unzip(input_layer, split_dim, shape[split_dim], num_splits)
return input_layer.with_sequence(splits)
def unzip(input_layer, split_dim=0, num_splits=2):
"""Unzips the head Tensor along the split_dim into num_splits Equal chunks.
Examples:
* `[1, 2, 3, 4] -> [1, 3], [2, 4]`
* `[[1, 1], [2, 2], [3, 3], [4, 4]] -> [[1, 1], [3, 3]], [[2, 2], [4, 4]]`
Args:
input_layer: The chainable object, supplied.
split_dim: The dimension to split along. Defaults to batch.
num_splits: The number of splits.
Returns:
A list of PrettyTensors.
Raises:
ValueError: If split_dim is out of range or isn't divided evenly by
num_splits.
"""
shape = input_layer.shape
_check_split_dims(num_splits, split_dim, shape)
splits = functions.unzip(input_layer, split_dim, shape[split_dim],
num_splits)
return input_layer.with_sequence(splits)