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

aclnnBinaryCrossEntropyWithLogits

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

每个算子分为两段式接口,必须先调用“aclnnBinaryCrossEntropyWithLogitsGetWorkspaceSize”接口获取入参并根据流程计算所需workspace大小,再调用“aclnnBinaryCrossEntropyWithLogits”接口执行计算。

  • aclnnStatus aclnnBinaryCrossEntropyWithLogitsGetWorkspaceSize(const aclTensor *self, const aclTensor *target, const aclTensor *weightOptional, const aclTensor *posWeightOptional, int64_t reduction, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)
  • aclnnStatus aclnnBinaryCrossEntropyWithLogits(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)

功能描述

  • 算子功能: 计算输入logits与标签target之间的BCELoss损失。

  • 计算公式:

单标签场景:

(self,target)=L={l1,...,ln}T\ell(self, target) = L = \{l_{1},..., l_{n}\}^{T} n=weightn[targeynlog(σ(selfn))+(1targetn)log(1σ(selfn))]\ell_{n} = -weight_{n}[targey_{n} \cdot log(\sigma(self_{n})) + (1 - target_{n}) \cdot log(1 - \sigma(self_{n}))] (self,target)={L,if reduction=nonemean(L),if reduction=meansum(L),if reduction=sum\ell(self, target) = \begin{cases} L, & if\ reduction = none\\ mean(L), & if\ reduction = mean\\ sum(L), & if\ reduction = sum\\ \end{cases}

多标签场景:

c(self,target)=Lc={l1,c,...,ln,c}T\ell_c(self, target) = L_c = \{l_{1,c},..., l_{n,c}\}^{T} n,c=weightn,c[pos_weightn,ctargeyn,clog(σ(selfn,c))+(1targetn,c)log(1σ(selfn,c))]\ell_{n,c} = -weight_{n,c}[pos\_weight_{n,c} \cdot targey_{n,c} \cdot log(\sigma(self_{n,c})) + (1 - target_{n,c}) \cdot log(1 - \sigma(self_{n,c}))]

aclnnBinaryCrossEntropyWithLogitsGetWorkspaceSize

  • 参数说明:

    • self(const aclTensor *, 计算输入): 连接层输出,Device侧的aclTensor,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。
    • target(const aclTensor *, 计算输入): lable标签值,Device侧的aclTensor,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),shape需要与self一致,支持非连续的Tensor数据格式支持ND。
    • weightOptional(const aclTensor *, 计算输入): 二分交叉熵权重,Device侧的aclTensor,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),shape需要能够broadcast到target,支持非连续的Tensor数据格式支持ND。
    • posWeightOptional(const aclTensor *, 计算输入): 各类的正类权重,Device侧的aclTensor,数据类型支持FLOAT16、FLOAT、BFLOAT16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持),shape需要能够broadcast到target,支持非连续的Tensor数据格式支持ND。
    • reduction(int64_t, 计算输入): 输出结果计算方式,Host侧的整型值,数据类型支持INT64,0代表None不操作;1代表Mean求损失函数均值;2代表Sum求损失函数的和。
    • out(aclTensor*, 计算输出): 输出误差,Device侧的aclTensor,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),且数据类型与target一致,如果reduction = None,shape与self一致,其他情况shape为[1],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

第一段接口完成入参校验,出现以下场景时报错:
161001(ACLNN_ERR_PARAM_NULLPTR):1. 传入的self或out为空指针。
161002(ACLNN_ERR_PARAM_INVALID):1. self、target、weightOptional和posWeightOptional的数据类型和数据格式不在支持的范围内。
                                 2. self和target维度不一致。
                                 3. weightOptional、posWeightOptional不能扩展成self/target形状。

aclnnBinaryCrossEntropyWithLogits

  • 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aclnnBinaryCrossEntropyWithLogitsGetWorkspaceSize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_binary_cross_entropy_with_logits.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根据自己的需要处理
  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> inputShape = {4, 2};
  std::vector<int64_t> targetShape = {4, 2};
  std::vector<int64_t> weightShape = {4, 2};
  std::vector<int64_t> posWeightShape = {4, 2};
  std::vector<int64_t> outShape = {4, 2};

  void* inputDeviceAddr = nullptr;
  void* targetDeviceAddr = nullptr;
  void* weightDeviceAddr = nullptr;
  void* posWeightDeviceAddr = nullptr;
  void* outDeviceAddr = nullptr;
  aclTensor* input = nullptr;
  aclTensor* target = nullptr;
  aclTensor* weight = nullptr;
  aclTensor* posWeight = nullptr;
  aclTensor* out = nullptr;

  std::vector<float> inputHostData = {0.1, 0.1, 0.2, 0.2, 0.3, 0.3, 0.4, 0.4};
  std::vector<float> targetHostData = {0.2, 0.2, 0.1, 0.1, 0.2, 0.2, 0.1, 0.1};
  std::vector<float> weightHostData = {0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5};
  std::vector<float> posWeightHostData = {0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5, 0.5};
  std::vector<float> outHostData = {0, 0, 0, 0, 0, 0, 0, 0};

  // 创建input aclTensor
  ret = CreateAclTensor(inputHostData, inputShape, &inputDeviceAddr, aclDataType::ACL_FLOAT, &input);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建target aclTensor
  ret = CreateAclTensor(targetHostData, targetShape, &targetDeviceAddr, aclDataType::ACL_FLOAT, &target);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建weight aclTensor
  ret = CreateAclTensor(weightHostData, weightShape, &weightDeviceAddr, aclDataType::ACL_FLOAT, &weight);
  CHECK_RET(ret == ACL_SUCCESS, return ret);
  // 创建posWeight aclTensor
  ret = CreateAclTensor(posWeightHostData, posWeightShape, &posWeightDeviceAddr, aclDataType::ACL_FLOAT, &posWeight);
  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);

  int64_t reduction = 0;

  uint64_t workspaceSize = 0;
  aclOpExecutor* executor;

  // aclnnBinaryCrossEntropyWithLogits接口调用示例
  // 3. 调用CANN算子库API,需要修改为具体的API名称
  // 调用aclnnBinaryCrossEntropyWithLogits第一段接口
  ret = aclnnBinaryCrossEntropyWithLogitsGetWorkspaceSize(input, target, weight, posWeight, reduction, out, &workspaceSize, &executor);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnBinaryCrossEntropyWithLogitsGetWorkspaceSize 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aclnnBinaryCrossEntropyWithLogits第二段接口
  ret = aclnnBinaryCrossEntropyWithLogits(workspaceAddr, workspaceSize, executor, stream);
  CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnBinaryCrossEntropyWithLogits 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 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(input);
  aclDestroyTensor(target);
  aclDestroyTensor(weight);
  aclDestroyTensor(posWeight);
  aclDestroyTensor(out);

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