ttnn.elu

ttnn.elu(input_tensor: ttnn.Tensor, *, alpha: float = 1, memory_config: ttnn.MemoryConfig = None, output_tensor: ttnn.Tensor = None, sub_core_grids: ttnn.CoreRangeSet = None) → ttnn.Tensor

Applies elu to input_tensor element-wise with alpha.

The alpha parameter for the ELU function

\[\mathrm{output\_tensor}_i = \verb|elu|(\mathrm{input\_tensor}_i, \verb|alpha|)\]

Parameters:

input_tensor (ttnn.Tensor) – the input tensor.

Keyword Arguments:

• alpha (float) – Defaults to 1.

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

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

• sub_core_grids (ttnn.CoreRangeSet, optional) – sub core grids for the operation. Defaults to None.

Returns:

ttnn.Tensor – the output tensor.

Note

Supported dtypes and layouts:

Dtypes	Layouts
BFLOAT16, BFLOAT8_B	TILE, ROW_MAJOR

Example

# Create a tensor with specific values
tensor = ttnn.from_torch(
    torch.tensor([[1, 2], [3, 4]], dtype=torch.bfloat16),
    dtype=ttnn.bfloat16,
    layout=ttnn.TILE_LAYOUT,
    device=device,
)

# Apply ELU activation function with alpha parameter
output = ttnn.elu(tensor, alpha=5)
logger.info(f"ELU: {output}")