ttnn.atan2

ttnn.atan2 = Operation(python_fully_qualified_name='ttnn.atan2', function=<ttnn._ttnn.operations.binary.atan2_t object>, preprocess_golden_function_inputs=<function default_preprocess_golden_function_inputs>, golden_function=<function _golden_function_atan2>, postprocess_golden_function_outputs=<function default_postprocess_golden_function_outputs>, is_cpp_operation=True, is_experimental=False)

Computes atan2 input_tensor_a and input_tensor_b and returns the tensor with the same layout as input_tensor_a

\[\mathrm{output\_tensor}_i = \arctan\left(\frac{\mathrm{input\_tensor\_a}_i}{\mathrm{input\_tensor\_b}_i}\right)\]
Parameters:
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, BFLOAT8_B

TILE

2, 3, 4

Input arguments for the atan2 function are in the format (y, x)

Example

>>> 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.atan2(tensor1, tensor2)