Documentation
Download
文档中心
aclnnTake

aclnnTake

📄 View Source Code

Applicable Products

Applicable ProductsSupported
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

  • Operator function: Treats the input self tensor as a one-dimensional array, uses the values of index as indices to extract elements from self, and outputs a tensor with the same shape as index.

  • Formula: Where the subscript i ranges from 0 to the number of elements in index minus 1.

outi=selfindex[i]out_{i} = self_{index[i]}

Function Prototype

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

  • aclnnStatus aclnnTakeGetWorkspaceSize(const aclTensor *self, const aclTensor *index, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)

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

aclnnTakeGetWorkspaceSize

  • Parameters

    • self(aclTensor*, input for computation): Target tensor to be converted, which is self in the formula. It is an aclTensor on the Device side. Supports non-contiguous tensor, data format supports ND, and data dimensions support higher than 8 scenarios.

      • Atlas training products: The data type supports UINT64, INT64, UINT32, FLOAT32, INT32, UINT16, INT16, FLOAT16, INT8, UINT8, DOUBLE, COMPLEX64, COMPLEX128, and BOOL.

      • Atlas A2 training products/Atlas A2 inference products, Atlas A3 training products/Atlas A3 inference products, Ascend 950PR/Ascend 950DT: The data type supports UINT64, INT64, UINT32, FLOAT32, INT32, UINT16, INT16, FLOAT16, BFLOAT16, INT8, UINT8, DOUBLE, COMPLEX64, COMPLEX128, and BOOL.

    • index(aclTensor*, input for computation): aclTensor on the Device side. The data type supports INT32 and INT64. Supports non-contiguous tensor, data dimensions not higher than 8, data format is ND, does not support negative indices, and the value range is within 0 to self.shape[dim] (inclusive of 0, exclusive of self.shape[dim]).

    • out(aclTensor*, compute output): Device-side aclTensor. The data type is the same as self, and the shape is consistent with index. Supports non-contiguous tensors. The data format is ND.

      • Atlas training products: The data types UINT64, INT64, UINT32, FLOAT32, INT32, UINT16, INT16, FLOAT16, INT8, UINT8, DOUBLE, COMPLEX64, COMPLEX128, and BOOL are supported.

      • Atlas A2 training products/Atlas A2 inference products, Atlas A3 training products/Atlas A3 inference products, Ascend 950PR/Ascend 950DT: The data types UINT64, INT64, UINT32, FLOAT32, INT32, UINT16, INT16, FLOAT16, BFLOAT16, INT8, UINT8, DOUBLE, COMPLEX64, COMPLEX128, and BOOL are supported.

    • workspaceSize(uint64_t*, output parameter): Returns the workspace size to be allocated on the Device side.

    • executor(aclOpExecutor**, output parameter): Returns the op executor, which contains the operator computation flow.

  • Return value

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

    The first-phase API completes input validation. The following scenarios trigger an error:
161001 (ACLNN_ERR_PARAM_NULLPTR): 1. The parameters self, index, and out are null pointers.
161002 (ACLNN_ERR_PARAM_INVALID): 1. The data types of the parameters self, index, and out are not within the supported range.
                                 2. The data formats of the parameters self, index, and out are private formats.
                                 3. The data types of self and out are inconsistent.
                                 4. The dimension of index is higher than 8.

aclnnTake

  • Parameters:

    • workspace(void *, input parameter): Address of the workspace memory applied on the Device side.

    • workspaceSize (uint64_t, input parameter): Device-side allocated workspace size, obtained from the first-phase API aclnnTakeGetWorkspaceSize.

    • executor(aclOpExecutor *, input parameter): Op executor, which contains the operator computation flow.

    • stream(aclrtStream, input parameter): Stream on which the task is executed.

  • Return value

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

Constraints

  • Deterministic computation:

    • aclnnTake 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_take.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, 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 apply for 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);
  // Call aclrtMemcpy to copy data from host to 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 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 device/stream. See the ACL API manual.
  // Fill in the 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);

  // Construct the input and output. Customize the construction based on the API.
  std::vector<int64_t> selfShape = {4, 2};
  std::vector<int64_t> indexShape = {2, 2};
  std::vector<int64_t> outShape = {2, 2};
  void* selfDeviceAddr = nullptr;
  void* indexDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* index = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {0, 1.123, -2.001, 303.45, 40009, -50.1234, 60.666, -7.6543};
  std::vector<int64_t> indexHostData = {4, 1, 0, 5};
  std::vector<float> outHostData = {0, 0, 0, 0};

  // Create the self aclTensor.
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the index aclTensor.
  ret = CreateAclTensor(indexHostData, indexShape, &indexDeviceAddr, aclDataType::ACL_INT64, &index);
  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);

  // 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 aclnnTake.
  ret = aclnnTakeGetWorkspaceSize(self, index, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnTakeGetWorkspaceSize 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 aclnnTake.
  ret = aclnnTake(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnTake 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 memory. 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);
  aclDestroyTensor(index);
  aclDestroyTensor(out);

  // 7. Release device resources.
  aclrtFree(selfDeviceAddr);
  aclrtFree(indexDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();

  return 0;
}
This site uses cookies to improve your browsing experience. By continuing to browse this site, you agree to our use of cookies.View Details