ttnn.add

ttnn.add = FastOperation(python_fully_qualified_name='ttnn.add', function=<ttnn._ttnn.operations.binary.add_t object>, preprocess_golden_function_inputs=<function default_preprocess_golden_function_inputs>, golden_function=<function _golden_function>, postprocess_golden_function_outputs=<function default_postprocess_golden_function_outputs>, is_cpp_operation=True, is_experimental=False)

Adds input_tensor_a to input_tensor_b and returns the tensor with the same layout as input_tensor_a

\[\mathrm{{output\_tensor}}_i = \mathrm{{input\_tensor\_a}}_i + \mathrm{{input\_tensor\_b}}_i\]

Args:

input_tensor_a (ttnn.Tensor): the input tensor. input_tensor_b (ttnn.Tensor or Number): the input tensor.

Keyword args:

memory_config (ttnn.MemoryConfig, optional): memory configuration for the operation. Defaults to None. dtype (ttnn.DataType, optional): data type for the output tensor. Defaults to None. output_tensor (ttnn.Tensor, optional): preallocated output tensor. Defaults to None. activations (List[str], optional): list of activation functions to apply to the output tensor: 'None' | 'relu'. Defaults to None. queue_id (int, optional): command queue id. Defaults to 0.

Returns:

ttnn.Tensor: the output tensor.

Supports broadcasting.

Note:

Supported dtypes, layouts, and ranks:

Dtypes	Layouts	Ranks
BFLOAT16, BFLOAT8_B, INT32	TILE	2, 3, 4

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.add(tensor1, tensor2/scalar)