ttnn.pow

ttnn.pow(input_tensor: ttnn.Tensor, float, exponent: float, int, ttnn.Tensor, *, memory_config: ttnn.MemoryConfig = None, output_tensor: ttnn.Tensor = None) → ttnn.Tensor

Perform element-wise pow operation on input_tensor with exponent.

\[\mathrm{output\_tensor}_i = (\mathrm{input\_tensor}_i ** \mathrm{exponent}_i)\]
Parameters:

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

  • exponent (float, int, ttnn.Tensor) – the exponent value.

Keyword Arguments:

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

  • output_tensor (ttnn.Tensor, optional) – preallocated output tensor. Defaults to None.

Returns:

ttnn.Tensor – the output tensor.

Binary elementwise operations, C=op(A,B), support input tensors A and B in row major and tile layout, in interleaved or sharded format (height, width or block sharded), in DRAM or L1. A and B are completely independent, and can have different tensor specs.

Broadcast of A and B operands is supported up to dimension 5 (DNCHW). Any dimensions of size 1 in either A or B will be expanded to match the other input, and data will be duplicated along that dimension. For example, if the shape of A is [2,1,1,32] and B is [1,16,8,1], the output shape will be [2,16,8,32]. The size of dimensions higher than 5 must match between A and B.

The output C also supports row major and tile layout, interleaved or sharded format (height, width or block sharded), in DRAM or L1. The tensor spec of C is independent of A and B, and can be explicitly set using the optional output tensor input; if not provided, the operation will attempt a best decision at an appropriate tensor spec. The dimensions of C, or equivalently the optional output tensor, must match the broadcast-matched size of A and B.

Performance considerations: Elementwise operations operate natively in tile format, tiled tensors are preferred as an input, and row-major tensors are tilized and untilized during the operation. L1 sharded layout is preferred, with no broadcast and matching tensor specs for A, B and C.

Note

Supported dtypes, layouts, and ranks:

Dtypes

Layouts

Ranks

BFLOAT16, BFLOAT8_B

TILE

2, 3, 4

When exponent is a Tensor, supported dtypes are: BFLOAT16, FLOAT32. Both input tensors should be of same dtype.

Example

# Create tensor and integer exponent
tensor = ttnn.from_torch(
    torch.tensor([[1, 2], [3, 4]], dtype=torch.bfloat16), layout=ttnn.TILE_LAYOUT, device=device
)
exponent = 2
output = ttnn.pow(tensor, exponent)
logger.info(f"Power with integer exponent result: {output}")

# Create tensor and float exponent
exponent = 2.5
output = ttnn.pow(tensor, exponent)
logger.info(f"Power with float exponent result: {output}")

# Create two tensors for exponentiation
tensor1 = ttnn.from_torch(
    torch.tensor([[1, 2], [3, 4]], dtype=torch.bfloat16), layout=ttnn.TILE_LAYOUT, device=device
)
tensor2 = ttnn.from_torch(
    torch.tensor([[1, 2], [3, 4]], dtype=torch.bfloat16), layout=ttnn.TILE_LAYOUT, device=device
)
output = ttnn.pow(tensor1, tensor2)
logger.info(f"Power with tensor exponent result: {output}")