ttnn.hardtanh_bw

ttnn.hardtanh_bw(grad_tensor: ttnn.Tensor, input_tensor: ttnn.Tensor, *, min: float | None = -1, max: float | None = 1, memory_config: ttnn.MemoryConfig | None = None) → List of ttnn.Tensor

Performs backward operations for hardtanh activation function on input_tensor, min, max with given grad_tensor.

Parameters:

• grad_tensor (ttnn.Tensor) – the input gradient tensor.

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

Keyword Arguments:

• min (float, optional) – Minimum value. Defaults to -1.

• max (float, optional) – Maximum value. Defaults to 1.

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

Returns:

List of ttnn.Tensor – the output tensor.

Note

Supported dtypes, layouts, and ranks:

Dtypes	Layouts	Ranks
BFLOAT16, BFLOAT8_B	TILE	2, 3, 4

Example

>>> grad_tensor = ttnn.from_torch(torch.tensor([[1, 2], [3, 4]], dtype=torch.bfloat16), layout=ttnn.TILE_LAYOUT, device=device)
>>> input = ttnn.from_torch(torch.tensor([[1, 2], [3, 4]], dtype=torch.bfloat16, requires_grad=True), layout=ttnn.TILE_LAYOUT, device=device)
>>> output = ttnn.hardtanh_bw(grad_tensor, input, min = -1, max = 1