ttnn.repeat_interleave

ttnn.repeat_interleave(input_tensor: ttnn.Tensor, repeats: number, dim: number, *, memory_config: ttnn.MemoryConfig = None) → ttnn.Tensor

Repeats elements of a tensor in the given dim.

Parameters:

  • input_tensor (ttnn.Tensor) – the input tensor.

  • repeats (number) – he number of repetitions for each element. repeats is broadcasted to fit the shape of the given axis.

  • dim (number) – the dimension to expand with the repetitions.

Keyword Arguments:

memory_config (ttnn.MemoryConfig, optional) – Memory configuration for the operation. Defaults to None.

Returns:

ttnn.Tensor – the output tensor.

Example

# Create a tensor to repeat interleave
input_tensor = ttnn.rand((10, 10), dtype=ttnn.bfloat16, layout=ttnn.Layout.TILE, device=device)

# Repeat interleave the tensor
repeated_tensor = ttnn.repeat_interleave(input_tensor, repeats=2, dim=0)
logger.info(
    "Repeat Interleave Tensor Shape:", repeated_tensor.shape
)  # Repeat Interleave Tensor Shape: Shape([20, 10])