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

aclnnGroupNormSiluV2

支持的产品型号

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

接口原型

每个算子分为两段式接口,必须先调用“aclnnGroupNormSiluV2GetWorkspaceSize”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“aclnnGroupNormSiluV2”接口执行计算。

  • aclnnStatus aclnnGroupNormSiluV2GetWorkspaceSize(const aclTensor *self, const aclTensor *gamma, const aclTensor *beta, int64_t group, double eps, bool activateSilu, aclTensor *out, aclTensor *meanOut, aclTensor *rstdOut, uint64_t *workspaceSize, aclOpExecutor **executor)
  • aclnnStatus aclnnGroupNormSiluV2(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, const aclrtStream stream)

功能描述

  • 接口功能:计算输入self的组归一化结果out,均值meanOut,标准差的倒数rstdOut,以及silu的输出。
  • 计算公式: 记 x=selfx=self, E[x]=xˉE[x] = \bar{x}代表xx的均值,Var[x]=1n1i=1n(xiE[x])2Var[x] = \frac{1}{n - 1} * \sum_{i=1}^n(x_i - E[x])^2代表xx的样本方差,则{out=xE[x]Var[x]+epsγ+βmeanOut=E[x]rstdOut=1Var[x]+eps\left\{ \begin{array} {rcl} out& &= \frac{x - E[x]}{\sqrt{Var[x] + eps}} * \gamma + \beta \\ meanOut& &= E[x]\\ rstdOut& &= \frac{1}{\sqrt{Var[x] + eps}}\\ \end{array} \right.

aclnnGroupNormSiluV2GetWorkspaceSize

  • 参数说明:

    • self(aclTensor*,计算输入):out计算公式中的xx,数据类型支持BFLOAT16、FLOAT16、FLOAT,维度需大于一维,数据格式支持ND,支持非连续的Tensor
    • gamma(aclTensor*,计算输入):可选参数,out计算公式中的γ\gamma,数据类型支持BFLOAT16、FLOAT16、FLOAT,维度为一维,元素数量需与输入selfself的第1维度相同,数据格式支持ND,支持非连续的Tensor
    • beta(aclTensor*,计算输入):可选参数,out计算公式中的β\beta,数据类型支持BFLOAT16、FLOAT16、FLOAT,维度为一维,元素数量需与输入selfself的第1维度相同,数据格式支持ND,支持非连续的Tensor
    • group(int,计算输入): INT32或者INT64常量,表示将输入selfself的第1维度分为group组。
    • eps(double,计算输入): DOUBLE常量,outrstdOut计算公式中的epseps值。
    • activateSilu(bool,计算输入): BOOL常量,是否支持silu计算。如果设置为true,则表示groupnorm计算后继续silu计算。
    • out(aclTensor*,计算输出): 输出张量,数据类型支持BFLOAT16、FLOAT16、FLOAT,数据类型和shape与selfself相同,数据格式支持ND,支持非连续的Tensor
    • meanOut(aclTensor*,计算输出): 输出张量,数据类型支持BFLOAT16、FLOAT16、FLOAT,数据类型与selfself相同,shape为(N, group),其中Nselfself的第0维度保持一致,数据格式支持ND,支持非连续的Tensor
    • rstdOut(aclTensor*,计算输出): 输出张量,数据类型支持BFLOAT16、FLOAT16、FLOAT,数据类型与selfself相同,shape为(N, group),其中Nselfself的第0维度保持一致,数据格式支持ND,支持非连续的Tensor
    • 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、meanOut、rstdOut是空指针时。
161002 ACLNN_ERR_PARAM_INVALID:1. self、gamma、beta、out、meanOut、rstdOut数据类型不在支持的范围之内。
                                2. out、meanOut、rstdOut的数据类型与self相同,gamma、beta与self可以不同。
                                3. gamma与beta的数据类型必须保持一致,且数据类型与self相同或者为FLOAT。
                                4. self维度不大于1。
                                5. self第1维度不能被group整除
                                6. eps小于等于0。
                                7. out的shape与self不同。
                                8. meanOut与rstdOut的shape不为(N, group),其中N为self第0维度值。
                                9. gamma不为1维或元素数量不等于输入self第1维度。
                                10. beta不为1维或元素数量不等于输入self第1维度。
                                11. group小于等于0。
                                12. self第0维小于等于0.
                                13. self第1维小于等于0.

aclnnGroupNormSiluV2

  • 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aclnnGroupNormSiluV2GetWorkspaceSize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_group_norm_silu.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acl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根据自己的需要处理
  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 = {2, 3, 4};
  std::vector<int64_t> gammaShape = {3};
  std::vector<int64_t> betaShape = {3};
  std::vector<int64_t> outShape = {2, 3, 4};
  std::vector<int64_t> meanOutShape = {2, 1};
  std::vector<int64_t> rstdOutShape = {2, 1};
  void* selfDeviceAddr = nullptr;
  void* gammaDeviceAddr = nullptr;
  void* betaDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  void* meanOutDeviceAddr = nullptr;
  void* rstdOutDeviceAddr = nullptr;
  aclTensor* self = nullptr;
  aclTensor* gamma = nullptr;
  aclTensor* beta = nullptr;
  aclTensor* out = nullptr;
  aclTensor* meanOut = nullptr;
  aclTensor* rstdOut = nullptr;
  std::vector<float> selfHostData = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0,
                                     13.0, 14.0, 15.0, 16.0, 17.0, 18.0, 19.0, 20.0, 21.0, 22.0, 23.0, 24.0};
  std::vector<float> gammaHostData = {2.0, 2, 2};
  std::vector<float> betaHostData = {2.0, 2, 2};
  std::vector<float> outHostData = {1.0, 2.0, 3.0, 4.0, 5.0, 6.0, 7.0, 8.0, 9.0, 10.0, 11.0, 12.0,
                                    13.0, 14.0, 15.0, 16.0, 17.0, 18.0, 19.0, 20.0, 21.0, 22.0, 23.0, 24.0};
  std::vector<float> meanOutHostData = {2.0, 2};
  std::vector<float> rstdOutHostData = {2.0, 2};

  int64_t group = 1;
  double eps = 0.00001;
  bool activateSilu = true;
  // 创建self aclTensor
  ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建gamma aclTensor
  ret = CreateAclTensor(gammaHostData, gammaShape, &gammaDeviceAddr, aclDataType::ACL_FLOAT, &gamma);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建beta aclTensor
  ret = CreateAclTensor(betaHostData, betaShape, &betaDeviceAddr, aclDataType::ACL_FLOAT, &beta);
  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);
  // 创建meanOut aclTensor
  ret = CreateAclTensor(meanOutHostData, meanOutShape, &meanOutDeviceAddr, aclDataType::ACL_FLOAT, &meanOut);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建rstdOut aclTensor
  ret = CreateAclTensor(rstdOutHostData, rstdOutShape, &rstdOutDeviceAddr, aclDataType::ACL_FLOAT, &rstdOut);
  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aclnnGroupNormSiluV2第一段接口
  ret = aclnnGroupNormSiluV2GetWorkspaceSize(self, gamma, beta, group, eps, activateSilu, out, meanOut, rstdOut, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnGroupNormSiluV2GetWorkspaceSize 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aclnnGroupNormSiluV2第二段接口
  ret = aclnnGroupNormSiluV2(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnGroupNormSiluV2 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> outResultData(size, 0);
  ret = aclrtMemcpy(outResultData.data(), outResultData.size() * sizeof(outResultData[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("outResultData[%ld] is: %f\n", i, outResultData[i]);
  }

  size = GetShapeSize(meanOutShape);
  std::vector<float> meanResultData(size, 0);
  ret = aclrtMemcpy(meanResultData.data(), meanResultData.size() * sizeof(meanResultData[0]), meanOutDeviceAddr, 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("meanResultData[%ld] is: %f\n", i, meanResultData[i]);
  }

  size = GetShapeSize(rstdOutShape);
  std::vector<float> rstdResultData(size, 0);
  ret = aclrtMemcpy(rstdResultData.data(), rstdResultData.size() * sizeof(rstdResultData[0]), rstdOutDeviceAddr, 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("rstdResultData[%ld] is: %f\n", i, rstdResultData[i]);
  }

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

  // 7. 释放device资源,需要根据具体API的接口定义修改
  aclrtFree(selfDeviceAddr);
  aclrtFree(gammaDeviceAddr);
  aclrtFree(betaDeviceAddr);
  aclrtFree(outDeviceAddr);
  aclrtFree(meanOutDeviceAddr);
  aclrtFree(rstdOutDeviceAddr);

  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