ttnn.isposinf

ttnn.isposinf(input_tensor: ttnn.Tensor, *, memory_config: ttnn.MemoryConfig | None = None) → ttnn.Tensor

Applies isposinf to input_tensor element-wise.

\[\begin{split}isposinf(\\mathrm{input\\_tensor}_i)\end{split}\]
Args:

  • input_tensor

Keyword Args:

  • memory_config (Optional[ttnn.MemoryConfig]): Memory configuration for the operation.

Example:

>>> tensor = ttnn.from_torch(torch.tensor((1, 2), dtype=torch.bfloat16), device=device)
>>> output = ttnn.isposinf(tensor)
Input Tensor 0

min_rank

max_rank

dtypes

layouts

can_be_on_device

can_be_on_cpu

can_be_scalar

is_optional

2

4

{DataType::BFLOAT16, DataType::BFLOAT8_B}

{Layout::ROW_MAJOR, Layout::TILE}

true

false

false

false