Cos

Applicability

Function Usage

Performs element-wise cosine computation using the following formula:

The Taylor's Formula of Cos(x) is as follows:

Prototype

• Pass the temporary space through the sharedTmpBuffer input parameter.
  • All or part of the source operand tensors are involved in computation.

    1 2	template <typename T, bool isReuseSource = false, const CosConfig& config = defaultCosConfig> __aicore__ inline void Cos(const LocalTensor<T>& dstTensor, const LocalTensor<T>& srcTensor, const LocalTensor<uint8_t>& sharedTmpBuffer, const uint32_t calCount)

  • All source operand tensors are involved in computation.

    1 2	template <typename T, bool isReuseSource = false, const CosConfig& config = defaultCosConfig> __aicore__ inline void Cos(const LocalTensor<T>& dstTensor, const LocalTensor<T>& srcTensor, const LocalTensor<uint8_t>& sharedTmpBuffer)

• Allocate the temporary space through the API framework.
  • All or part of the source operand tensors are involved in computation.

    1 2	template <typename T, bool isReuseSource = false, const CosConfig& config = defaultCosConfig> __aicore__ inline void Cos(const LocalTensor<T>& dstTensor, const LocalTensor<T>& srcTensor, const uint32_t calCount)

  • All source operand tensors are involved in computation.

    1 2	template <typename T, bool isReuseSource = false, const CosConfig& config = defaultCosConfig> __aicore__ inline void Cos(const LocalTensor<T>& dstTensor, const LocalTensor<T>& srcTensor)

Due to the complex mathematical computation involved in the internal implementation of this API, additional temporary space is required to store intermediate variables generated during computation. The temporary space can be passed by developers through the sharedTmpBuffer input parameter or allocated through the API framework.

• When the sharedTmpBuffer input parameter is used for passing the temporary space, the tensor serves as the temporary space. In this case, the API framework is not required for temporary space allocation. This enables developers to manage the sharedTmpBuffer space and reuse the buffer after calling the API, so that the buffer is not repeatedly allocated and deallocated, improving the flexibility and buffer utilization.
• When the API framework is used for temporary space allocation, developers do not need to allocate the space, but must reserve the required size for the space.

If sharedTmpBuffer is used, developers must allocate space for the tensor. If the API framework is used, developers must reserve the temporary space. To obtain the size of the temporary space (BufferSize) to be reserved, use the API provided in GetCosMaxMinTmpSize.

Parameters

1
2
3
4
5
6
7

struct CosConfig {
  CosAlgo algo = CosAlgo::POLYNOMIAL_APPROXIMATION;
};
enum class CosAlgo {
  POLYNOMIAL_APPROXIMATION = 0,
  RADIAN_REDUCTION,
};

Returns

None

Restrictions

• For the Atlas 350 Accelerator Card, when the polynomial fitting algorithm POLYNOMIAL_APPROXIMATION is used in the template parameter config, ensure that the input source data must be within the value range of [–65504.0, 65504.0].
• For the following products, the input source data must be within the value range of [–65504.0, 65504.0].
  • Atlas A3 training product/Atlas A3 inference product
  • Atlas A2 training product/Atlas A2 inference product
  • Atlas inference product AI Core
• The source operand address must not overlap the destination operand address.
• The address of sharedTmpBuffer must not overlap the addresses of the source operand and destination operand.
• For details about the operand address alignment requirements, see General Address Alignment Restrictions.

Example

For a complete operator sample, see Cos operator sample.

1
2
3
4
5
6
7
8

// dstLocal: tensor for storing the computation result
// srcLocal: input tensor involved in computation
// sharedTmpBuffer: temporary buffer for storing intermediate variables during complex internal computation
// The input tensor length is 1,024, the input data type of the operator is half, and the number of actually computed data elements is 512.
AscendC::Cos(dstLocal, srcLocal, sharedTmpBuffer, 512);
constexpr AscendC::CosAlgo algo = AscendC::CosAlgo::RADIAN_REDUCTION;
constexpr AscendC::CosConfig config = { algo };
AscendC::Cos<half, false, config>(dstLocal, srcLocal, sharedTmpBuffer, 512);

1
2

Input (srcLocal): [0.00            0.01            0.02             ...  5.10            5.11]
Output (dstLocal): [1.00000000e+00 9.99949992e-01 9.99800026e-01 ... 3.77977639e-01 3.87216508e-01])