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

aclnnExpm1&aclnnInplaceExpm1

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。

接口原型

  • aclnnExpm1和aclnnInplaceExpm1实现相同的功能,使用区别如下,请根据自身实际场景选择合适的算子。

    • aclnnExpm1:需新建一个输出张量对象存储计算结果。
    • aclnnInplaceExpm1:无需新建输出张量对象,直接在输入张量的内存中存储计算结果。
  • 每个算子分为两段式接口,必须先调用“aclnnExpm1GetWorkspaceSize”或者“aclnnInplaceExpm1GetWorkspaceSize”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“aclnnExpm1”或者“aclnnInplaceExpm1”接口执行计算。

    • aclnnStatus aclnnExpm1GetWorkspaceSize(const aclTensor *self, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)
    • aclnnStatus aclnnExpm1(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, const aclrtStream stream)
    • aclnnStatus aclnnInplaceExpm1GetWorkspaceSize(aclTensor *selfRef, uint64_t *workspaceSize, aclOpExecutor **executor)
    • aclnnStatus aclnnInplaceExpm1(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, const aclrtStream stream)

功能描述

  • 算子功能:以输入的self为指数,计算自然常数e的幂,并对指数计算结果进行减1计算。对于self取值较小的场景,提供比直接用公式计算结果更高的精度。
  • 计算公式:outi=eselfi1out_i = {e}^{self_i} - 1

aclnnExpm1GetWorkspaceSize

  • 参数说明:

    • self(aclTensor*,计算输入):公式中输入self,数据类型支持INT64、BOOL、FLOAT、BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持)、FLOAT16,shape需要与out一致。支持非连续的Tensor数据格式支持ND。
    • out(aclTensor*,计算输出):公式中输入out,数据类型支持FLOAT、FLOAT16、FLOAT64、BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持),shape需要与self一致。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self、out是空指针。
    161002(ACLNN_ERR_PARAM_INVALID): 1. self和out的数据类型不在支持的范围之内。
                                     2. self的数据类型无法转换为指定输出out的类型。
                                     3. self和out的shape不一致。

aclnnExpm1

  • 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aclnnExpm1GetWorkspaceSize获取。
    • executor(aclOpExecutor*,入参):op执行器,包含了算子计算流程。
    • stream(const 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

aclnnInplaceExpm1GetWorkspaceSize

  • 参数说明:

    • selfRef(aclTensor*,计算输入|计算输出):输入输出tensor,即公式中输入selfout,数据类型支持FLOAT、FLOAT16、BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持)。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selfRef是空指针。
    161002(ACLNN_ERR_PARAM_INVALID): 1. selfRef的数据类型不在支持的范围之内。

aclnnInplaceExpm1

  • 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aclnnInplaceExpm1GetWorkspaceSize获取。
    • executor(aclOpExecutor*,入参):op执行器,包含了算子计算流程。
    • stream(const 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

aclnnExpm1示例代码:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_expm1.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) {
  // 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侧引擎
  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 == 0, 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> selfShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};
  void* selfDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* out = nullptr;
  std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  std::vector<float> outHostData = {1, 1, 1, 1, 0, 0, 0, 0};

  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  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
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // 调用aclnnExpm1第一段接口
  ret = aclnnExpm1GetWorkspaceSize(self, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnExpm1GetWorkspaceSize 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aclnnExpm1第二段接口
  ret = aclnnExpm1(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnExpm1 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(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("result[%ld] is: %f\n", i, resultData[i]);
  }

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

  // 7. 释放device资源,需要根据具体API的接口定义修改
  aclrtFree(selfDeviceAddr);
  aclrtFree(outDeviceAddr);
  if (workspaceSize > 0) {
    aclrtFree(workspaceAddr);
  }
  aclrtDestroyStream(stream);
  aclrtResetDevice(deviceId);
  aclFinalize();
  return 0;
}

aclnnInplaceExpm1示例代码:

#include <iostream>
#include <vector>
#include "acl/acl.h"
#include "aclnnop/aclnn_expm1.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) {
  // 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侧引擎
  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 == 0, 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> selfRefShape = {4, 2};
  void* selfRefDeviceAddr = nullptr;
  aclTensor* selfRef = nullptr;
  std::vector<float> selfRefHostData = {0, 1, 2, 3, 4, 5, 6, 7};
  // 创建self aclTensor
  ret = CreateAclTensor(selfRefHostData, selfRefShape, &selfRefDeviceAddr, aclDataType::ACL_FLOAT, &selfRef);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 3. 调用CANN算子库API,需要修改为具体的API
  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;
  // 调用aclnnInplaceExpm1第一段接口
  ret = aclnnInplaceExpm1GetWorkspaceSize(selfRef, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceExpm1GetWorkspaceSize 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aclnnInplaceExpm1第二段接口
  ret = aclnnInplaceExpm1(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnInplaceExpm1 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(selfRefShape);
  std::vector<float> resultData(size, 0);
  ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), selfRefDeviceAddr, 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("result[%ld] is: %f\n", i, resultData[i]);
  }
  // 6. 释放aclTensor和aclScalar,需要根据具体API的接口定义修改
  aclDestroyTensor(selfRef);

  // 7. 释放device资源,需要根据具体API的接口定义修改
  aclrtFree(selfRefDeviceAddr);
  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