concatenate

Graph.concatenate(tensors, axis, *, name=None)

Concatenate a list of tensors along a specified dimension.

Parameters:

tensors (list[np.ndarray or Tensor]) – The list of tensors that you want to concatenate. All of them must have the same shape in all dimensions except axis.
axis (int) – The dimension along which you want to concatenate the tensors.
name (str or None, optional) – The name of the node.

Returns:

The concatenated tensor.

Return type:

Tensor

Notes

This node only concatenates on an existing axis, it does not create new axes. If you want to stack along a new axis or concatenate scalars, add a new axis to the tensors with [None].

Examples

>>> x = np.array([[1, 2, 3], [4, 5, 6]])
>>> y = np.array([[7, 8, 9]])

Concatenate x and y along their first dimension.

>>> graph.concatenate(tensors=[x, y], axis=0, name="node_0")
<Tensor: name="node_0", operation_name="concatenate", shape=(3, 3)>
>>> result = qctrl.functions.calculate_graph(graph=graph, output_node_names=["node_0"])
>>> result.output["node_0"]["value"]
array([[1., 2., 3.],
       [4., 5., 6.],
       [7., 8., 9.]])

Concatenate two x arrays along their second dimension.

>>> graph.concatenate(tensors=[x, x], axis=1, name="node_1")
<Tensor: name="node_1", operation_name="concatenate", shape=(2, 6)>
>>> result = qctrl.functions.calculate_graph(graph=graph, output_node_names=["node_1"])
>>> result.output["node_1"]["value"]
array([[1., 2., 3., 1., 2., 3.],
       [4., 5., 6., 4., 5., 6.]])