aclnnNanMedianDim

📄 View Source Code

Applicable Products

Function

• Description: After ignoring NaN values, returns the median and its index along the specified dimension of the tensor.

• Example:

  • Example 1:

    When keepDim is True, set the size of the corresponding dimension to 1; when it is False, remove the corresponding dimension.
    Assume the shape of self is [2, 3, 4], dim = 1, and keepDim is true; then the output shape is [2, 1, 4].
    Assume the shape of self is [2, 3, 4], dim = 1, and keepDim is false; then the output shape is [2, 4].

  • Example 2:

    Example of the output shape
    If the input is
    self = tensor([[1, float('nan'), 3, 2],[-1, float('nan'), 3, 2]]) with shape [2, 4]
    dim = 0
    keepDim = true
    then the output is
    valuesOut = tensor([[-1., float('nan'),  3.,  2.]]) with shape [1, 4]
    indicesOut = tensor([[1, 0, 0, 0]]) with shape [1, 4]

  • Example 3:

    If the input is
    self = tensor([[1, float('nan'), 3, 2],[-1, float('nan'), 3, 2]]) with shape [2, 4]
    dim = 0
    keepDim = false
    then the output is
    valuesOut = tensor([-1., float('nan'),  3.,  2.]) with shape [4]
    indicesOut = tensor([1, 0, 0, 0]) with shape [4]

  • Example 4:

    If the input is
    self = tensor([[1, float('nan'), 3, 2],[-1, float('nan'), 3, 2]]) with shape [2, 4]
    dim = 1
    keepDim = false
    then the output is
    valuesOut = tensor([2, 2]) with shape [2]
    indicesOut = tensor([3, 3]) with shape [2]

Function Prototype

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

aclnnStatus aclnnNanMedianDimGetWorkspaceSize(
  const aclTensor* self,
  int64_t          dim,
  bool             keepDim,
  aclTensor*       valuesOut,
  aclTensor*       indicesOut,
  uint64_t*        workspaceSize,
  aclOpExecutor**  executor)

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

aclnnNanMedianDimGetWorkspaceSize

• Parameters:

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
  self	Input		-	FLOAT, FLOAT16, UINT8, INT8, INT16, INT32, INT64, BFLOAT16.	ND	0-8	√
  dim	Input	Specified dimension.	Value range: [-self.dim(), self.dim() - 1].		-	-	-
  keepDim	Input	Whether to retain the input tensor's dimension in the output tensor.	If true, sets the size of the corresponding dimension to 1; if false, removes the corresponding dimension.
  valuesOut	Output	The median value.	If keepDim is true, the shape must match the shape of self in all dimensions except dim, with size 1 in the dim dimension. If keepDim is false, the shape must match the shape of self excluding the dim dimension. Non-contiguous tensors are supported. The data format supports ND.	FLOAT, FLOAT16, UINT8, INT8, INT16, INT32, INT64, BFLOAT16.	ND	0–8	√
  indicesOut	Output	Index of the median.		INT64	ND	0-8	√
  workspaceSize	Output	Returns the workspace size to be requested on the Device side.	-	-	-	-	-
  executor	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

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

• 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 Value	Error Code	Description
  ACLNN_ERR_PARAM_NULLPTR	161001	Indicates that self, valuesOut, or indicesOut is a null pointer.
  ACLNN_ERR_PARAM_INVALID	161002	The data type of self, valuesOut, or indicesOut is not within the supported range.
  		The data types of self and valuesOut are different.
  		The value range of dim exceeds [-self.dim(), self.dim() - 1].
  		The dimension of self, valuesOut, or indicesOut exceeds 8.
  		The size of self in the dim dimension cannot be 0.
  		When keepDim is true, the dimension of valuesOut or indicesOut is inconsistent with that of self.
  		When keepDim is false, the dimension of valuesOut or indicesOut is not one less than that of self.
  		When keepDim is true, the shape of valuesOut or indicesOut is inconsistent with that of self in sizes other than dim.
  		When keepDim is true, the size of the shape of valuesOut or indicesOut on dim is not 1.
  		When keepDim is false, the shape of valuesOut or indicesOut is inconsistent with the shape of self except for dim.

aclnnNanMedianDim

• Parameters

• Return Value

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

Constraints

When the data type of self is not FLOAT, FLOAT16, or BFLOAT16, an excessively large tensor size may cause the operator execution to time out (aicpu error, with the error reason being [aicpu timeout]). The maximum size limits for specific types (strongly dependent on the actual remaining memory of the machine) are as follows:

- INT64: 150000000

- UINT8, INT8, INT16, INT32: 725000000

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 <cmath>
#include "acl/acl.h"
#include "aclnnop/aclnn_median.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 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 memory on the device side.
  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 data from the host side to the device memory.
  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 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);
  // Handle the check based on your requirements.
  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 = {4, 2};
  std::vector<int64_t> valuesOutShape = {2};
  std::vector<int64_t> indicesOutShape = {2};
  void* selfDeviceAddr = nullptr;
  void* valuesOutDeviceAddr = nullptr;
  void* indicesOutDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* valuesOut = nullptr;
  aclTensor* indicesOut = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7, NAN};
  std::vector<float> valuesOutHostData = {0, 0};
  std::vector<int64_t> indicesOutHostData = {0, 0};
  int64_t dim = 0;
  bool keepDim = false;
  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the valuesOut aclTensor.
  ret = CreateAclTensor(valuesOutHostData, valuesOutShape, &valuesOutDeviceAddr, aclDataType::ACL_FLOAT, &valuesOut);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the indicesOut aclTensor.
  ret = CreateAclTensor(indicesOutHostData, indicesOutShape, &indicesOutDeviceAddr, aclDataType::ACL_INT64, &indicesOut);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. Call the CANN operator library API. Replace with the specific API.
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // Call the first-phase API of aclnnNanMedianDim.
  ret = aclnnNanMedianDimGetWorkspaceSize(self, dim, keepDim, valuesOut, indicesOut, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnNanMedianDimGetWorkspaceSize 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 aclnnNanMedianDim.
  ret = aclnnNanMedianDim(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnNanMedianDim 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 memory. Modify based on the specific API definition.
  auto size = GetShapeSize(valuesOutShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), valuesOutDeviceAddr,
                    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("valuesOut[%ld] is: %f\n", i, resultData[i]);
  }

  std::vector<int64_t> indicesData(size, 0);
  ret = aclrtMemcpy(indicesData.data(), indicesData.size() * sizeof(indicesData[0]), indicesOutDeviceAddr,
                    size * sizeof(int64_t), 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("indicesOut[%ld] is: %ld\n", i, indicesData[i]);
  }

  // 6. Release aclTensor and aclScalar. Modify based on the specific API definition.
  aclDestroyTensor(self);
  aclDestroyTensor(valuesOut);
  aclDestroyTensor(indicesOut);

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