aclnnReduceNansum

📄 View source code

Supported Products

Function

Description: Returns the sum of a specified dimension of the input tensor after NaN in the tensor is processed to 0.

Prototype

Each operator has two-phase API calls. First, aclnnReduceNansumGetWorkspaceSize is called to obtain the workspace size required for computation and the executor that contains the operator computation process. Then, aclnnReduceNansum is called to perform computation.

• aclnnStatus aclnnReduceNansumGetWorkspaceSize(const aclTensor* self, const aclIntArray* dim, bool keepDim, aclDataType dtype, aclTensor* out, uint64_t* workspaceSize, aclOpExecutor** executor)
• aclnnStatus aclnnReduceNansum(void* workspace, uint64_t workspaceSize, aclOpExecutor* executor, aclrtStream stream)

aclnnReduceNansumGetWorkspaceSize

• Parameters:

  • self (aclTensor*, computation input): input tensor. The data type can be FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, UINT8, BOOL, or BFLOAT16, and must be convertible from/to that of out (for details, see Conversion Relationship). Non-contiguous tensors are supported. The data format can be ND.

  • dim (aclIntArray*, computation input): dimension involved in computation. The value range is [–self.dim(), self.dim()]. If the length of the dim array is 0, ReduceNansum computation is performed on all axes. The data type can be INT64.

  • keepDim (bool, computation input): Whether to retain the dimension to be reduced in the output tensor.

  • dtype (aclDataType, computation input): required data type of the output tensor. The data type can be FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, UINT8, BOOL, or BFLOAT16.

  • out (aclTensor*, computation output): output tensor. The data type can be FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, UINT8, BOOL, or BFLOAT16, and must be convertible from/to that of self (for details, see Conversion Relationship). Non-contiguous tensors are supported. The data format can be ND.

  • workspaceSize (uint64_t*, output): size of the workspace to be allocated on the device.

  • executor (aclOpExecutor**, output): operator executor, containing the operator computation process.

• Returns:

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

   The first-phase API implements input parameter verification. The following errors may be returned:
 161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The passed `self`, `dim`, or `out` is a null pointer.
 161002 (ACLNN_ERR_PARAM_INVALID): 1. The data type of `self` or `out` is not supported.
                                   2. The dimensions specified in the `dim` array exceed the range of dimensions of the input tensor.
                                   3. The axes specified by `dim` are duplicate.
                                   4. The shape of `self` or `out` exceeds eight dimensions.
                                   5. The data type of `dtype` is different from that of `out`.
                                   6. The shape of `out` does not match the actual shape.

aclnnReduceNansum

• Parameters:

  • workspace (void*, input): memory address of the workspace to be allocated on the device.

  • workspaceSize (uint64_t, input): size of the workspace to be allocated on the device, which is obtained by calling the first-phase API aclnnReduceNansumGetWorkspaceSize.

  • executor (aclOpExecutor*, input): operator executor, containing the operator computation process.

  • stream (aclrtStream, input): stream for executing the task.

• Returns:

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

Constraints

• Deterministic computation:
  • aclnnReduceNansum defaults to a deterministic implementation.

Example

The following example is for reference only. For details, see Compilation and Running Sample.

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_reduce_nansum.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) Initialize resources.
  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 memory on the device.
  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);
  // Call aclrtMemcpy to copy the data on the host to the memory on the device.
  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 contiguous tensors.
  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 aclCreateTensor 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 main() {
  // 1. (Boilerplate) Initialize the device and stream. For details, see the ACL API manual.
  // Set the device ID 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 based on the API definition.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> outShape = {2};
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclIntArray* dim = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<float> outHostData = {0, 0};
  std::vector<int64_t> dimData = {0};
  bool keepDim = false;
  auto dtype = aclDataType::ACL_FLOAT;
  // Create a self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create an out aclTensor.
  ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create a dim aclIntArray.
  dim = aclCreateIntArray(dimData.data(), 1);
  CHECK_RET(dim != nullptr, return ret);

// 3. Call the CANN operator library API. Modify the API name to the actual one.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnReduceNansum.
  ret = aclnnReduceNansumGetWorkspaceSize(self, dim, keepDim, dtype, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnReduceNansumGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // Allocate device memory based on the computed workspaceSize.
  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 aclnnReduceNansum.
  ret = aclnnReduceNansum(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnReduceNansum failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Wait until the task execution is 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 to the host. Modify the code based on the 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 the aclTensor and aclIntArray. Modify the code based on the API definition.
  aclDestroyTensor(self);
  aclDestroyIntArray(dim);
  aclDestroyTensor(out);

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