aclnnMeanV2

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 mean of a tensor along specified dimensions.
Formula:

Y = \frac{1}{|\text{dim}|} \sum_{\text{dim}} X

Function Prototype

Each operator is divided into a two-phase API. The "aclnnMeanV2GetWorkspaceSize" API must be called first to obtain the workspace size required for computation and the executor that encapsulates the operator computation flow. Then, the "aclnnMeanV2" API is called to perform the computation.

aclnnStatus aclnnMeanV2GetWorkspaceSize(
  const aclTensor*     self,
  const aclIntArray*   dim,
  bool                 keepDim,
  bool                 noopWithEmptyAxes,
  aclTensor*           out,
  uint64_t*            workspaceSize,
  aclOpExecutor**      executor)

aclnnStatus aclnnMeanV2(
  void*          workspace,
  uint64_t       workspaceSize,
  aclOpExecutor* executor,
  aclrtStream    stream)

aclnnMeanV2GetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Shape	Non-contiguous Tensor
self (aclTensor*)	Input	X in the formula.	When the input is an empty tensor, the output type cannot be complex types COMPLEX64 and COMPLEX128.	FLOAT, FLOAT16, BFLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, COMPLEX64, COMPLEX128	ND	0-8	√
dim (aclIntArray*)	Input	Input array.	The value range is [-r, r-1], where r is the dimension of the input data.	INT64	-	-	-
keepDim (bool)	Input	Specifies whether to retain the reduced dimension.	-	BOOL	-	-	-
noopWithEmptyAxes (bool)	Input	Defines the behavior when dim is set to [].	When set to false, dim being [] reduces all axes. When set to true, dim being [] retains all axes, and the output tensor is identical to the input tensor.	BOOL	-	-	-
out (aclTensor*)	Output	Y in the formula.	The data type must be the same as self.	FLOAT, FLOAT16, BFLOAT16, DOUBLE, INT32, INT64, INT16, INT8, UINT8, COMPLEX64, COMPLEX128	ND	0-8	√
workspaceSize (uint64_t*)	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
executor（aclOpExecutor**）	Output	Returns the op executor, which contains the operator computation flow.	-	-	-	-	-

Atlas inference products and Atlas training products: The BFLOAT16 data type is not supported.

Return value

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

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

Return Value	Error Code	Description
ACLNN_ERR_PARAM_NULLPTR	161001	The passed self, out, or dim is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data type of self or out is not within the supported range.
		The data format of self or out is not within the supported range.
		A dimension in the dim array exceeds the dimension range of the input tensor.
		The dim array contains duplicate elements.

aclnnMeanV2

Parameters:

Parameter	Input/Output	Description
workspace	Input	The workspace memory address applied for on the Device side.
workspaceSize	Input	The workspace size applied for on the Device side, which is obtained through the first-phase API aclnnMeanV2GetWorkspaceSize.
executor	Input	Operator executor that contains the operator computation flow.
stream	Input	Specifies the stream that executes the task.

Return value

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

Constraints

Deterministic computation:
- aclnnMeanV2 defaults to a deterministic implementation.

Example

For sample compilation and execution, see Compile and Run Samples.

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_mean.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 shape_size = 1;
  for (auto i : shape) {
    shape_size *= i;
  }
  return shape_size;
}

int Init(int32_t deviceId, aclrtStream* stream) {
  // Boilerplate: 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 host-side data to device-side memory using 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);

  // Calculate 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];
  }

  // Call the aclCreateTensor API to create an aclTensor.
  *tensor = aclCreateTensor(shape.data(), shape.size(), dataType, strides.data(), 0, aclFormat::ACL_FORMAT_ND,
                            shape.data(), shape.size(), *deviceAddr);
  return 0;
}

int PrepareInputAndOutput(
    std::vector<int64_t>& selfShape, std::vector<int64_t>& outShape, void** selfDeviceAddr, aclTensor** self, aclIntArray** dim,
    void** outDeviceAddr, aclTensor** out)
{
    std::vector<int64_t> selfHostData = {2, 3, 5, 8, 4, 12, 6, 7};
    std::vector<int64_t> outHostData = {2, 3, 5, 8};
    std::vector<int64_t> dimData = {1, 2};

    // Create the self aclTensor.
    auto ret = CreateAclTensor(selfHostData, selfShape, selfDeviceAddr, aclDataType::ACL_INT64, self);
    CHECK_RET(ret == ACL_SUCCESS, return ret);
    // Create the dim aclIntArray.
    *dim = aclCreateIntArray(dimData.data(), 1);
    CHECK_RET(ret == ACL_SUCCESS, return false);
    // Create the out aclTensor.
    ret = CreateAclTensor(outHostData, outShape, outDeviceAddr, aclDataType::ACL_INT64, out);
    CHECK_RET(ret == ACL_SUCCESS, return ret);

    return ACL_SUCCESS;
}

void ReleaseTensorAndIntArray(aclTensor* self, aclIntArray* dim, aclTensor* out)
{
    aclDestroyTensor(self);
    aclDestroyIntArray(dim);
    aclDestroyTensor(out);
}

void ReleaseDevice(
    void* selfDeviceAddr, void* outDeviceAddr, uint64_t workspaceSize, void* workspaceAddr, aclrtStream stream,
    int32_t deviceId)
{
    aclrtFree(selfDeviceAddr);
    aclrtFree(outDeviceAddr);
    if (workspaceSize > 0) {
        aclrtFree(workspaceAddr);
    }
    aclrtDestroyStream(stream);
    aclrtResetDevice(deviceId);
    aclFinalize();
}

int main() {
  // 1. (Boilerplate) Initialize the device/stream. See the acl API manual.
  // Set deviceId based on your actual device.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  // Handle the check result as needed.
  CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
  // 2. Construct the input and output. Customize the construction based on the API.
  std::vector<int64_t> selfShape = {2, 2, 2};
  std::vector<int64_t> outShape = {2, 2};

  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclIntArray* dim = nullptr;
  aclTensor* out = nullptr;

  ret = PrepareInputAndOutput(selfShape, outShape, &selfDeviceAddr, &self, &dim, &outDeviceAddr, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API.
  uint64_t workspaceSize = 0;
  bool keepdim = false;
  bool noopWithEmptyAxes = true;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnMeanV2.
  ret = aclnnMeanV2GetWorkspaceSize(self, dim, keepdim, noopWithEmptyAxes, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMeanV2GetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Apply for device memory based on the workspaceSize calculated 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 aclnnMeanV2.
  ret = aclnnMeanV2(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnMeanV2 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 side memory to the host side. 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(float),
                    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. Modify based on the specific API definition.
  ReleaseTensorAndIntArray(self, dim, out);

  // 7. Release device resources. Modify based on the specific API definition.
  ReleaseDevice(selfDeviceAddr, outDeviceAddr, workspaceSize, workspaceAddr, stream, deviceId);

  return 0;
}