aclnnEmbedding

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

Maps a dataset to a vector space, thereby quantizing the data. The two-dimensional embedding weight tensor is weight (with m+1 rows and n columns). For any input index tensor indices (e.g., 1 row and 3 columns), the output out is a tensor with 3 rows and n columns, as shown below:

\begin{aligned} & weight = \begin{bmatrix} x_{01} & x_{02} & ... & x_{0n}\\ x_{11} & x_{12} & ... & x_{1n}\\ x_{21} & x_{22} & ... & x_{2n}\\ ...\\ x_{m1} & x_{m2} & ... & x_{mn}\\ \end{bmatrix}\\ & indices = \begin{bmatrix} 0,2,m \end{bmatrix}\\ & out = \begin{bmatrix} x_{01} & x_{02} & ... & x_{0n}\\ x_{21} & x_{22} & ... & x_{2n}\\ x_{m1} & x_{m2} & ... & x_{mn}\\ \end{bmatrix}\\ \end{aligned}

Function Prototype

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

aclnnStatus aclnnEmbeddingGetWorkspaceSize(
 const aclTensor *weight,
 const aclTensor *indices,
 const aclTensor *out,
 uint64_t        *workspaceSize,
 aclOpExecutor  **executor)

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

aclnnEmbeddingGetWorkspaceSize

Parameters:

Parameter	Input/Output	Description	Instruction	Data Type	Data Format	Dimension (Shape)	Non-Contiguous Tensor
weight	Input	Embedding matrix.	-	BFLOAT16、FLOAT、FLOAT16、INT64、INT32、INT16、INT8、UINT8、BOOL、DOUBLE、COMPLEX64、COMPLEX128	ND	2	√
indices	Input	Indices pointing to the embedding matrix.	The value ranges from 0 to weight.shape[0] (inclusive of 0, exclusive of weight.shape[0]).	INT64, INT32	ND	-	√
out	Output	Result after retrieval from the embedding matrix.	dtype must be the same as weight.	BFLOAT16, FLOAT, FLOAT16, INT64, INT32, INT16, INT8, UINT8, BOOL, DOUBLE, COMPLEX64, COMPLEX128	ND	-	√
workspaceSize	Output	Size of the workspace to be requested on the Device side.	-	-	-	-	-
executor	Output	Returns the operator executor, which contains the operator computation flow.	-	-	-	-	-

Atlas inference products, Atlas training products, Atlas 200I/500 A2 inference products: do not support the BFLOAT16 data type.

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 input weight, indices, or out is a null pointer.
ACLNN_ERR_PARAM_INVALID	161002	The data type or data format of weight, indices, or out is not within the supported range.
		The shapes of indices and out do not satisfy the matching relationship.
		The maximum dimension of the input or output exceeds 8.
		The operator reports an error when indices contain negative values, and outputs random values when out-of-bounds access occurs in the forward direction.

aclnnEmbedding

Parameters:

Parameter	Input/Output	Description
workspace	Input	Memory address of the workspace applied for on the Device side.
workspaceSize	Input	Size of the workspace applied for on the Device side, obtained through the first-phase API aclnnEmbeddingGetWorkspaceSize.
executor	Input	Operator executor, which contains the operator computation flow.
stream	Input	Specifies the stream that executes the task.

Return value

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

Constraints

Deterministic computation:
aclnnEmbedding 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_embedding.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-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 to the device memory by calling 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);

  // Compute 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];
  }

  // Create an aclTensor by calling aclCreateTensor.
  *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);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);

  // 2. Construct the input and output. Customize the construction based on the API interface.
  std::vector<int64_t> weightShape = {10, 3};
  std::vector<int64_t> indicesShape = {4};
  std::vector<int64_t> outShape = {4, 3};
  void* weightDeviceAddr = nullptr;
  void* indicesDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* weight = nullptr;
  aclTensor* indices = nullptr;
  aclTensor* out = nullptr;

  std::vector<float> weightHostData = {1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23,
                                      24, 25, 26, 27, 28, 29, 30};
  std::vector<int64_t> indicesHostData = {1, 2, 3, 4};
  std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0};

  // Create the weight aclTensor.
  ret = CreateAclTensor(weightHostData, weightShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // Create the indices aclTensor.
  ret = CreateAclTensor(indicesHostData, indicesShape, &indicesDeviceAddr, aclDataType::ACL_INT64, &indices);
  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 aclnnEmbedding.
  ret = aclnnEmbeddingGetWorkspaceSize(weight, indices, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEmbeddingGetWorkspaceSize 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 aclnnEmbedding.
  ret = aclnnEmbedding(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEmbedding failed. ERROR: %d\n", ret); return ret);

  // 4. (Boilerplate) Wait for the task execution to complete synchronously.
  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 this 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 resultData from device to host failed. ERROR: %d\n", ret);
            return ret);
  for (int64_t i = 0; i < size; i++) {
    LOG_PRINT("resultData[%ld] is: %f\n", i, resultData[i]);
  }

  // Release the aclTensor. Adjust based on the specific API definition.
  aclDestroyTensor(weight);
  aclDestroyTensor(indices);
  aclDestroyTensor(out);

  // Release the device resource. Adjust based on the specific API definition.
  aclrtFree(weightDeviceAddr);
  aclrtFree(indicesDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}