nvidia.dali.fn.cat — NVIDIA DALI (original) (raw)

nvidia.dali.fn.cat(__input_0, /, *__input_, axis=0, axis_name=None, bytes_per_sample_hint=[0], preserve=False, device=None, name=None)#

Joins the input tensors along an existing axis.

The shapes of the inputs must match in all dimensions except the concatenation axis.

Supported backends

• ‘cpu’
• ‘gpu’

Parameters:

• __input_0¶ (TensorList) – Input to the operator.
• __input_¶[ 1..998 ] (TensorList , optional) – This function accepts up to 998 optional positional inputs

Keyword Arguments:

• axis¶ (int, optional, default = 0) –
  Axis along which the input tensors are concatenated.
  Accepted range is [-ndim, ndim-1]. Negative indices are counted from the back.
• axis_name¶ (str , optional) –
  Name of the axis along which the tensors are concatenated.
  This argument is mutually exclusive with axis. This argument requires that at least one input has a non-empty layout and that all non-empty input layouts match.
• bytes_per_sample_hint¶ (int or list of int, optional, default = [0]) –
  Output size hint, in bytes per sample.
  If specified, the operator’s outputs residing in GPU or page-locked host memory will be preallocated to accommodate a batch of samples of this size.
• preserve¶ (bool, optional, default = False) – Prevents the operator from being removed from the graph even if its outputs are not used.