aclnnReduceLogSum
支持的产品型号
Atlas 推理系列产品 。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“aclnnReduceLogSumGetWorkspaceSize”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“aclnnReduceLogSum”接口执行计算。
aclnnStatus aclnnReduceLogSumGetWorkspaceSize(const aclTensor* data, const aclIntArray* axes, bool keepDims, bool noopWithEmptyAxes, aclTensor* reduce, uint64_t* workspaceSize, aclOpExecutor** executor)
aclnnStatus aclnnReduceLogSum(void* workspace, uint64_t workspaceSize, aclOpExecutor* executor, aclrtStream stream)
功能描述
算子功能:返回给定维度中输入张量每行的和再取对数。
aclnnReduceLogSumGetWorkspaceSize
参数说明:
- data(aclTensor*, 计算输入):表示参与计算的目标张量,维度小于8维,Device侧的aclTensor,支持非连续的Tensor,数据类型支持FLOAT16、FLOAT32,数据格式支持ND。
- axes(aclIntArray*, 计算输入):指定计算维度,Host侧的aclIntArray,数据类型支持INT64,取值范围为[-self.dim(), self.dim()-1]。
- keepDims(bool, 计算输入):指定是否在输出张量中保留输入张量的维度,Host侧的BOOL值。
- noopWithEmptyAxes(bool, 计算输入):指定axes为空时的行为:false即对所有轴进行计算;true即不进行计算,输出张量等于输入张量,Host侧的BOOL值。
- reduce(aclTensor*, 计算输出):表示计算后的结果,维度小于8维,Device侧的aclTensor,支持非连续的Tensor,数据类型支持FLOAT16、FLOAT32,需与data一致,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data、axes或reduce是空指针。
161002(ACLNN_ERR_PARAM_INVALID): 1. data或reduce的数据类型不在支持的范围之内。
2. reduce shape与实际不匹配。
3. axes数组中的维度超出输入tensor的维度范围。
4. axes数组中的轴重复。
aclnnReduceLogSum
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aclnnReduceSumGetWorkspaceSize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
无
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_reduce_log_sum.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/context/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 AscendCL 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> dataShape = {4, 2};
std::vector<int64_t> outShape = {2};
void* dataDeviceAddr = nullptr;
void* outDeviceAddr = nullptr;
aclTensor* data = nullptr;
aclIntArray* axes = nullptr;
aclTensor* out = nullptr;
std::vector<float> dataHostData = {0, 1, 2, 3, 4, 5, 6, 7};
std::vector<float> outHostData = {0, 0};
std::vector<int64_t> axesData = {0};
bool keepDims = false;
bool noopWithEmptyAxes = false;
// 创建data aclTensor
ret = CreateAclTensor(dataHostData, dataShape, &dataDeviceAddr, aclDataType::ACL_FLOAT, &data);
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axes aclIntArray
axes = aclCreateIntArray(axesData.data(), 1);
CHECK_RET(axes != nullptr, 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aclnnReduceLogSum第一段接口
ret = aclnnReduceLogSumGetWorkspaceSize(data, axes, keepDims, noopWithEmptyAxes, out, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnReduceLogSumGetWorkspaceSize 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aclnnReduceLogSum第二段接口
ret = aclnnReduceLogSum(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnReduceLogSum 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aclIntArray,需要根据具体API的接口定义修改
aclDestroyTensor(data);
aclDestroyIntArray(axes);
aclDestroyTensor(out);
// 7. 释放device资源,需要根据具体API的接口定义修改
aclrtFree(dataDeviceAddr);
aclrtFree(outDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}