ttnn.threshold

ttnn.threshold(input_tensor: ttnn.Tensor, threshold: float, value: float, *, memory_config: ttnn.MemoryConfig = None) → ttnn.Tensor

Performs threshold function on input_tensor, threshold, value.

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

Parameters:

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

• threshold (float) – Threshold value.

• value (float) – Replacing value.

Keyword Arguments:

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

Returns:

ttnn.Tensor – the output tensor.

Note

Supported dtypes, layouts, and ranks:

Dtypes	Layouts	Ranks
BFLOAT16	TILE	2, 3, 4

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,
)
threshold = 1.0
value = 10.0

# Apply threshold function
output = ttnn.threshold(tensor, threshold, value)
logger.info(f"Threshold: {output}")