下载
中文
注册
我要评分
文档获取效率
文档正确性
内容完整性
文档易理解
在线提单
论坛求助
昇腾小AI

aclnnEmbedding

支持的产品型号

  • Atlas 推理系列产品。
  • Atlas 训练系列产品。
  • Atlas A2训练系列产品/Atlas 800I A2推理产品。
  • Atlas 200/500 A2推理产品。

接口原型

每个算子分为两段式接口,必须先调用“aclnnEmbeddingGetWorkspaceSize”接口获取计算所需workspace大小以及包含了算子计算流程的执行器,再调用“aclnnEmbedding”接口执行计算。

  • 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)

功能描述

算子功能:把数据集合映射到向量空间,进而将数据进行量化。embedding的二维权重张量为weight(m+1行,n列),对于任意输入索引张量indices(如1行3列),输出out是一个3行n列的张量,如下所示:

weight=[x01x02...x0nx11x12...x1nx21x22...x2n...xm1xm2...xmn]indices=[0,2,m]out=[x01x02...x0nx21x22...x2nxm1xm2...xmn]\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}

aclnnEmbeddingGetWorkspaceSize

  • 参数说明:

    • weight(aclTensor*, 计算输入):Device侧的aclTensor,数据类型支持BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持)、FLOAT、FLOAT16、INT64、INT32、INT16、INT8、UINT8、BOOL、DOUBLE、COMPLEX64、COMPLEX128,支持非连续的Tensor数据格式支持ND, shape支持2维。
    • indices(aclTensor*, 计算输入):Device侧的aclTensor。数据类型支持INT64、INT32,支持非连续的Tensor数据格式支持ND。indices中的索引数据不支持越界。
    • out(aclTensor*, 计算输出):Device侧的aclTensor,数据类型支持BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持)、FLOAT、FLOAT16、INT64、INT32、INT16、INT8、UINT8、BOOL、DOUBLE、COMPLEX64、COMPLEX128,dtype需要与weight相同,支持非连续的Tensor数据格式支持ND。
    • workspaceSize(uint64_t*, 出参):返回需要在Device侧申请的workspace大小。
    • executor(aclOpExecutor**, 出参):返回op执行器,包含了算子计算流程。
  • 返回值:

    aclnnStatus:返回状态码,具体参见aclnn返回码

    161001(ACLNN_ERR_PARAM_NULLPTR):1. 传入的weight、indices、或out是空指针。
    161002(ACLNN_ERR_PARAM_INVALID):1. weight、indices和out的数据类型和数据格式不在支持的范围之内。
                                     2. indices和out的shape不满足匹配关系。
                                     3. 输入、输出的最大维度超过8。
                                     4. 当indices中有负值时算子报错,正向越界时输出随机值。

aclnnEmbedding

  • 参数说明:

    • workspace(void *, 入参):在Device侧申请的workspace内存地址。
    • workspaceSize(uint64_t, 入参):在Device侧申请的workspace大小,由第一段接口aclnnEmbeddingGetWorkspaceSize获取。
    • executor(aclOpExecutor *, 入参):op执行器,包含了算子计算流程。
    • stream(aclrtStream, 入参):指定执行任务的 AscendCL Stream流。
  • 返回值:

    aclnnStatus:返回状态码,具体参见aclnn返回码

约束与限制

调用示例

示例代码如下,仅供参考,具体编译和执行过程请参考编译与运行样例

#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) {
  // 固定写法,AscendCL初始化
  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);
  // 调用aclrtMalloc申请device侧内存
  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);
  // 调用aclrtMemcpy将host侧数据拷贝到device侧内存上
  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);

  // 计算连续tensor的strides
  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];
  }

  // 调用aclCreateTensor接口创建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. (固定写法)device/stream初始化,参考AscendCL对外接口列表
  // 根据自己的实际device填写deviceId
  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. 构造输入与输出,需要根据API的接口自定义构造
  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};

  // 创建weight aclTensor
  ret = CreateAclTensor(weightHostData, weightShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建indices aclTensor
  ret = CreateAclTensor(indicesHostData, indicesShape, &indicesDeviceAddr, aclDataType::ACL_INT64, &indices);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建out aclTensor
  ret = CreateAclTensor(outHostData, outShape, &outDeviceAddr, aclDataType::ACL_FLOAT, &out);
  CHECK_RET(ret == ACL_SUCCESS, return ret);

  // 3. 调用CANN算子库API,需要修改为具体的Api名称
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // 调用aclnnEmbedding第一段接口
  ret = aclnnEmbeddingGetWorkspaceSize(weight, indices, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEmbeddingGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
  // 根据第一段接口计算出的workspaceSize申请device内存
  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);
  }
  // 调用aclnnEmbedding第二段接口
  ret = aclnnEmbedding(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEmbedding failed. ERROR: %d\n", ret); return ret);

  // 4. (固定写法)同步等待任务执行结束
  ret = aclrtSynchronizeStream(stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);

  // 5. 获取输出的值,将device侧内存上的结果拷贝至host侧,需要根据具体API的接口定义修改
  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]);
  }

  // 6. 释放aclTensor,需要根据具体API的接口定义修改
  aclDestroyTensor(weight);
  aclDestroyTensor(indices);
  aclDestroyTensor(out);

  // 7. 释放device资源,需要根据具体API的接口定义修改
  aclrtFree(weightDeviceAddr);
  aclrtFree(indicesDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}
搜索结果
找到“0”个结果

当前产品无相关内容

未找到相关内容,请尝试其他搜索词