Pad

Product	Supported
Atlas A3 training products/Atlas A3 inference products	√
Atlas A2 training products/Atlas A2 inference products	√
Atlas 200I/500 A2 inference products	x
Atlas inference products	√
Atlas training products	√

Pads dimensions of an input tensor based on paddings. The padding value is 0.

const aclTensor* Pad(const aclTensor* self, const aclTensor* paddings, aclOpExecutor* executor)

Parameter	Input/Output	Description
self	Input	Input tensor to be padded. The data type can be FLOAT16, FLOAT, INT16, UINT16, INT32, BFLOAT16, INT8, or INT64. The data format can be ND. NOTE: BFLOAT16 and INT8 apply to the following product models: Atlas A2 training products/Atlas A2 inference products Atlas A3 training products/Atlas A3 inference products
paddings	Input	Padding sizes of dimensions of the input tensor. The shape is [self.dim, 2]. The data type can be INT32 or INT64, and the data format can be ND.
executor	Input	Operator executor, containing the operator computation process.

A tensor padded with 0s.

None

//Call l0op::Pad to perform dimension padding on self.
l0op::Pad(self, paddingsTensor, executor);

Parent topic: Basic Tensor Operation APIs