aclnnKthvalue

📄 View Source Code

Applicable Products

Function

Returns the kth smallest value and its index of the input tensor along a specified dimension.

Function Prototype

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

aclnnStatus aclnnKthvalueGetWorkspaceSize(
 const aclTensor*     self,
 int64_t              k,
 int64_t              dim,
 bool                 keepdim,
 aclTensor*           valuesOut,
 aclTensor*           indicesOut,
 uint64_t*            workspaceSize,
 aclOpExecutor**      executor)

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

aclnnKthvalueGetWorkspaceSize

• Parameters

  Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-contiguous Tensor
  self	Input	-	-	BFLOAT16, FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, UINT8, UINT16, UINT32, UINT64	ND	1-8	√
  k	Input	An integer on the host, indicating the kth smallest value along the specified dimension.	The value range is [0, self.size(dim)].	INT64	-	-	-
  dim	Input	An integer on the host side, indicating the dimension along which to select from the input tensor.	The value range is [-self.dim(), self.dim()).	INT64	-	-	-
  keepdim	Input.	A bool type value that indicates whether the output tensor retains the dim.	True: The valuesOut and indicesOut tensors have the same size as self. False: The dim is compressed, and the resulting tensor has one fewer dimension than input.	BOOL	-	-	-
  valuesOut	Output	Device-side aclTensor. The data type is consistent with self.	The sorted axis of the shape is 1, and the non-sorted axes are consistent with self.	Consistent with self.	ND	1-8	√
  indicesOut	Output	aclTensor on the device side, representing the indices of the kth smallest values along the dim dimension in the original input tensor.	The shape has a sorting axis of 1, and the non-sorting axes are consistent with self.	INT64	ND	1-8	√
  workspaceSize	Output	Returns the workspace size to be applied on the Device side.	-	-	-	-	-
  Executor	Output	Returns the op executor, which includes the operator computation stream.	-	-	-	-	-

  • Atlas inference products and Atlas training products: FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, and UINT8 data types are supported.

  • Atlas A2 training products/Atlas A2 inference products and Atlas A3 training products/Atlas A3 inference products: BFLOAT16, FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, and UINT8 data types are supported.

  • Ascend 950PR/Ascend 950DT: Data type support: BFLOAT16, FLOAT16, FLOAT32, INT8, INT16, INT32, INT64, UINT8, UINT16, UINT32, UINT64.

• 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	The passed 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, or the shapes do not match each other.
  		The value of dim is not within the dimension range of the input tensor self.
  		k is less than 0 or k is greater than the size of input self along the dim dimension.

aclnnKthvalue

• Parameters

  Parameter	Input/Output	Description
  workspace	Input	Address of the workspace memory applied on the Device side.
  workspaceSize	Input	Size of the workspace applied on the Device side, obtained through the first-phase API aclnnKthvalueGetWorkspaceSize.
  executor	Input	Operator executor that includes 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:

  • aclnnKthvalue defaults to a deterministic implementation.

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_kthvalue.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. Initializes 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 apply for device-side 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 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 the 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. Refer to the ACL API manual.
  // Enter the actual device ID.
  int32_t deviceId = 0;
  aclrtStream stream;
  auto ret = Init(deviceId, &stream);
  // Handle the check based on your needs.
  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, 4};
  std::vector<int64_t> outShape = {2, 1};
  void* selfDeviceAddr = nullptr;
  void* valuesOutDeviceAddr = nullptr;
  void* indicesOutDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  uint64_t dim = 1;
  uint64_t k = 2;
  bool keepdim = true;
  aclTensor* valuesOut = nullptr;
  aclTensor* indicesOut = nullptr;
  std::vector<float> selfHostData = {0.0, 1.1, 2, 3, 4, 5, 6, 7};
  std::vector<float> valuesOutHostData = {0.0, 0};
  std::vector<int64_t> indicesOutHostData = {0, 0};
  // 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, outShape, &valuesOutDeviceAddr, aclDataType::ACL_FLOAT, &valuesOut);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the indicesOut aclTensor.
  ret = CreateAclTensor(indicesOutHostData, outShape, &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 aclnnKthvalue first-phase API.
  ret = aclnnKthvalueGetWorkspaceSize(self, k, dim, keepdim, valuesOut, indicesOut, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnKthvalueGetWorkspaceSize 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 aclnnKthvalue second-phase API.
  ret = aclnnKthvalue(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnKthvalue failed. ERROR: %d\n", ret); return ret);
  // 4. (Boilerplate) Synchronize and 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> valuesData(size, 0);
  ret = aclrtMemcpy(valuesData.data(), valuesData.size() * sizeof(valuesData[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("valuesResult[%ld] is: %f\n", i, valuesData[i]);
  }
  std::vector<float> indicesData(size, 0);
  ret = aclrtMemcpy(indicesData.data(), indicesData.size() * sizeof(indicesData[0]), indicesOutDeviceAddr, 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("indicesResult[%ld] is: %f\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;
}