aclnnLinalgVectorNorm

Applicable Products

Product	Supported
Ascend 950PR/Ascend 950DT	√
Atlas A3 training products/Atlas A3 inference products	√
Atlas A2 training products/Atlas A2 inference products	√
Atlas 200I/500 A2 inference products	√
Atlas inference products	×
Atlas training products	√

Function

Description: Computes the vector norm of the input tensor.
Formula: In the formula, n is the size of the dimension specified by the dims parameter.

out=(\sum_{i=1}^{n}|self_{i}|^{ord})^{1/ord}

Function Prototype

Each operator is divided into a two-phase API. The "aclnnLinalgVectorNormGetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnLinalgVectorNorm" API is called to perform the computation.

aclnnStatus aclnnLinalgVectorNormGetWorkspaceSize(
  const aclTensor    *self,
  const aclScalar    *ord,
  const aclIntArray   *dims,
  bool                keepDims,
  const aclDataType   dtype,
  aclTensor          *out,
  uint64_t           *workspaceSize,
  aclOpExecutor     **executor)

aclnnStatus aclnnLinalgVectorNorm(
  void           *workspace,
  uint64_t        workspaceSize,
  aclOpExecutor  *executor,
  aclrtStream     stream)

aclnnLinalgVectorNormGetWorkspaceSize

Parameters

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
self (aclTensor*)	Input	Represents the input tensor for computation. Corresponds to self in the formula.	Supports empty tensors.	FLOAT32, FLOAT16, BFLOAT16	ND	0-8	√
ord (aclScalar*)	Input	Indicates the type of the norm. Corresponds to ord in the formula.	-	FLOAT32	-	-	-
dims (aclIntArray*)	Input	Specifies the dimensions for computing the norm.	Supports [-N, N-1], where N is the dimension of self. Elements in dims must not be duplicated. Defaults to None, in which case self is flattened before computation.	INT64	-	-	-
keepDims (bool)	Input	Determines whether the output tensor retains the axes specified by the dims parameter.	-	-	-	-	-
dtype (aclDataType)	Input	Specifies the data type used for computing self. It must be consistent with the dtype of out.	Converts self to the type specified by dtype before computation. The dtype and the data type of self must satisfy the following constraints: When the data type of self is FLOAT32, dtype can only be FLOAT32. When the data type of self is FLOAT16, dtype can be FLOAT or FLOAT16. When the data type of self is BFLOAT16, dtype can be FLOAT or BFLOAT16.	FLOAT32, FLOAT16, BFLOAT16	-	-	-
out (aclTensor*)	Output	Computes the output tensor. Corresponds to `out` in the formula.	If `keepDims` is `true`, the shape of all dimensions except the dimension specified by `dim` (whose size is 1) must be consistent with `self`. If `keepDims` is `false`, the dimensions of the reduced axes are not preserved, and the shapes of the remaining dimensions must be consistent with `self`.	FLOAT32, FLOAT16, BFLOAT16	ND	0-8	√
workspaceSize (uint64_t*)	Output	Returns the workspace size to be allocated on the Device side.	-	-	-	-	-
executor (aclOpExecutor**)	Output	Returns the op executor, which contains the operator computation flow.	-	-	-	-	-

Ascend 950PR/Ascend 950DT: The ord parameter additionally supports the values 2147483647.0, -2147483648.0, Inf, and -Inf.
Atlas inference products, Atlas training products, Atlas 200I/500 A2 inference products: do not support the BFLOAT16 data type.

Return Value

aclnnStatus: return code. For details, see aclnn Return Code.

The first-phase API performs input parameter validation and returns an error in the following scenarios:

Return Code	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The self, ord, dims, or out parameter passed in is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002
		dims exceeds the range [-N, N-1] (where N is the dimension of self).
		The data type of dtype is inconsistent with the data type of out.
		Values in dims are duplicated.
		The value of ord is not within the supported range.
		The shape of self or out exceeds 8 dimensions.
		The shape of out is not equal to the shape derived from self, dim, and keepDim.

aclnnLinalgVectorNorm

Parameters

Parameter	Input/Output	Description
workspace	Input	Address of the workspace memory applied for on the Device side.
workspaceSize	Input	Size of the workspace applied for on the Device side, obtained through the first-phase API aclnnLinalgVectorNormGetWorkspaceSize.
Executor	Input	The operator executor that contains the operator computation flow.
Stream	Input	Specifies the stream for task execution.

Return Value

aclnnStatus: return code. For details, see aclnn Return Code.

Constraints

Deterministic computation
- aclnnLinalgVectorNorm defaults to a non-deterministic implementation. Deterministic computation can be enabled through aclrtCtxSetSysParamOpt.

Example

The following provides example code for reference only. For details about compilation and running, see Compile and Run Samples.

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_linalg_vector_norm.h"

#define CHECK_RET(cond, return_expr) \
    do {                             \
        if (!(cond)) {               \
            return_expr;             \
        }                            \
    } while (0)

#define LOG_PRINT(message, ...)         \
    do {                                \
        printf(message, ##__VA_ARGS__); \
    } while (0)

int64_t GetShapeSize(const std::vector<int64_t>& shape)
{
    int64_t shapeSize = 1;
    for (auto i : shape) {
        shapeSize *= i;
    }
    return shapeSize;
}

int Init(int32_t deviceId, aclrtStream* stream)
{
    // Boilerplate for resource initialization.
    auto ret = aclInit(nullptr);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclInit failed. ERROR: %d\n", ret); return ret);
    ret = aclrtSetDevice(deviceId);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSetDevice failed. ERROR: %d\n", ret); return ret);
    ret = aclrtCreateStream(stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtCreateStream failed. ERROR: %d\n", ret); return ret);
    return 0;
}

template <typename T>
int CreateAclTensor(
    const std::vector<T>& hostData, const std::vector<int64_t>& shape, void** deviceAddr, aclDataType dataType,
    aclTensor** tensor)
{
    auto size = GetShapeSize(shape) * sizeof(T);
    // Call aclrtMalloc to allocate device memory.
    auto ret = aclrtMalloc(deviceAddr, size, ACL_MEM_MALLOC_HUGE_FIRST);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMalloc failed. ERROR: %d\n", ret); return ret);
    // Copy data from the host to the device memory by calling aclrtMemcpy.
    ret = aclrtMemcpy(*deviceAddr, size, hostData.data(), size, ACL_MEMCPY_HOST_TO_DEVICE);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtMemcpy failed. ERROR: %d\n", ret); return ret);

    // Compute the strides of a contiguous tensor.
    std::vector<int64_t> strides(shape.size(), 1);
    for (int64_t i = shape.size() - 2; i >= 0; i--) {
        strides[i] = shape[i + 1] * strides[i + 1];
    }

    // Create an aclTensor by calling aclCreateTensor.
    *tensor = aclCreateTensor(
        shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND, shape.data(), shape.size(),
        *deviceAddr);
    return 0;
}

int main()
{
    // 1. (Boilerplate) Initialize the device/stream. See the ACL API manual.
    // Set deviceId based on the actual device.
    int32_t deviceId = 0;
    aclrtStream stream;
    auto ret = Init(deviceId, &stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

    // 2. Construct the input and output. Custom construction based on the API is required.
    std::vector<int64_t> selfShape = {4, 2};
    std::vector<int64_t> outShape = {1, 2};

    void* selfDeviceAddr = nullptr;
    void* outDeviceAddr = nullptr;
    aclTensor* self = nullptr;
    aclTensor* out = nullptr;
    aclScalar* ord = nullptr;
    aclIntArray* dims = nullptr;

    std::vector<float> selfHostData = {0.1, 1.1, 2.1, 3.1, 4.1, 5.1, 6.1, 7.1};
    std::vector<float> outHostData = {0.0, 0.0};
    std::vector<int64_t> dimData = {0};
    float pValue = 2.0f;
    bool keepDims = true;
    aclDataType dtype = aclDataType::ACL_FLOAT;

    // Create the self aclTensor.
    ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the out aclTensor.
    ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the ord aclScalar.
    ord = aclCreateScalar(&pValue, aclDataType::ACL_FLOAT);
    CHECK_RET(ord != nullptr, return ret);
    // Create the dim aclIntArray.
    dims = aclCreateIntArray(dimData.data(), 1);
    CHECK_RET(dims != nullptr, return ret);

    // 3. Call the CANN operator library API. Replace with the specific API name.
    uint64_t workspaceSize = 0;
    aclOpExecutor* executor;
    // Call the first-phase API of aclnnLinalgVectorNorm.
    ret = aclnnLinalgVectorNormGetWorkspaceSize(self, ord, dims, keepDims, dtype, out, &workspaceSize, &executor);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLinalgVectorNormGetWorkspaceSize failed. ERROR: %d\n", ret);
              return ret);
    // Apply for device memory based on the workspaceSize computed by the first-phase API.
    void* workspaceAddr = nullptr;
    if (workspaceSize > 0) {
        ret = aclrtMalloc(&workspaceAddr, workspaceSize, ACL_MEM_MALLOC_HUGE_FIRST);
        CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("allocate workspace failed. ERROR: %d\n", ret); return ret);
    }
    // Call the second-phase API of aclnnLinalgVectorNorm.
    ret = aclnnLinalgVectorNorm(workspaceAddr, workspaceSize, executor, stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnLinalgVectorNorm failed. ERROR: %d\n", ret); return ret);

    // 4. (Boilerplate) Synchronously wait for task execution to complete.
    ret = aclrtSynchronizeStream(stream);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

    // 5. Obtain the output value and copy the result from the device memory to the host. Modify based on the specific API definition.
    auto size = GetShapeSize(outShape);
    std::vector<float> resultData(size, 0);
    ret = aclrtMemcpy(
        resultData.data(), resultData.size() * sizeof(resultData[0]), outDeviceAddr, size * sizeof(resultData[0]),
        ACL_MEMCPY_DEVICE_TO_HOST);
    CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("copy result from device to host failed. ERROR: %d\n", ret); return ret);
    for (int64_t i = 0; i < size; i++) {
        LOG_PRINT("result[%ld] is: %f\n", i, resultData[i]);
    }

    // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
    aclDestroyTensor(self);
    aclDestroyIntArray(dims);
    aclDestroyScalar(ord);
    aclDestroyTensor(out);

    // 7. Release device resources. Modify based on the specific API definition.
    aclrtFree(selfDeviceAddr);
    aclrtFree(outDeviceAddr);
    if (workspaceSize > 0) {
        aclrtFree(workspaceAddr);
    }
    aclrtDestroyStream(stream);
    aclrtResetDevice(deviceId);
    aclFinalize();

    return 0;
}