aclnnForeachLogInplace【免费下载链接】ops-nn本项目是CANN提供的神经网络类计算算子库实现网络在NPU上加速计算。项目地址: https://gitcode.com/cann/ops-nn产品支持情况产品是否支持Ascend 950PR/Ascend 950DT√Atlas A3 训练系列产品/Atlas A3 推理系列产品×Atlas A2 训练系列产品/Atlas A2 推理系列产品×Atlas 200I/500 A2 推理产品×Atlas 推理系列产品×Atlas 训练系列产品×功能说明接口功能原地更新输入张量列表对输入张量列表中的每个张量逐元素求自然对数。计算公式$$ x [{x_0}, {x_1}, ... {x_{n-1}}]\ $$$$ x_i \ln(x_i) (i0,1,...n-1) $$函数原型每个算子分为两段式接口必须先调用“aclnnForeachLogInplaceGetWorkspaceSize”接口获取入参并根据计算流程计算所需workspace大小再调用“aclnnForeachLogInplace”接口执行计算。aclnnStatus aclnnForeachLogInplaceGetWorkspaceSize( const aclTensorList *x, uint64_t *workspaceSize, aclOpExecutor **executor)aclnnStatus aclnnForeachLogInplace( void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)aclnnForeachLogInplaceGetWorkspaceSize参数说明参数名输入/输出描述使用说明数据类型数据格式维度(shape)非连续TensorxaclTensorList*输入/输出表示进行自然对数运算的输入和输出张量列表对应公式中的x。支持空Tensor。该参数中所有Tensor的数据类型保持一致。FLOAT32、FLOAT16、BFLOAT16ND0-8√workspaceSizeuint64_t*输出返回需要在Device侧申请的workspace大小。-----executoraclOpExecutor**输出返回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返回码错误码描述ACLNN_ERR_PARAM_NULLPTR161001传入的x是空指针。ACLNN_ERR_PARAM_INVALID161002x的数据类型不在支持的范围之内。ACLNN_ERR_INNER_TILING_ERROR561002x中的Tensor的数据类型不一致。x中的Tensor维度超过8维。aclnnForeachLogInplace参数说明参数名输入/输出描述workspace输入在Device侧申请的workspace内存地址。workspaceSize输入在Device侧申请的workspace大小由第一段接口aclnnForeachLogInplaceGetWorkspaceSize获取。executor输入op执行器包含了算子计算流程。stream输入指定执行任务的Stream。返回值aclnnStatus返回状态码具体参见aclnn返回码。约束说明确定性计算aclnnForeachLogInplace默认确定性实现。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_foreach_log_inplace.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::vectorint64_t shape) { int64_t shapeSize 1; for (auto i : shape) { shapeSize * i; } return shapeSize; } int Init(int32_t deviceId, aclrtStream* stream) { // 固定写法资源初始化 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::vectorT hostData, const std::vectorint64_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侧数据复制到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::vectorint64_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 API手册 // 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::vectorint64_t selfShape1 {2, 3}; std::vectorint64_t selfShape2 {1, 3}; void* input1DeviceAddr nullptr; void* input2DeviceAddr nullptr; aclTensor* input1 nullptr; aclTensor* input2 nullptr; std::vectorfloat input1HostData {1.0, 2.0, 3.0, 4.0, 5.0, 6.0}; std::vectorfloat input2HostData {7.0, 8.0, 9.0}; // 创建input1 aclTensor ret CreateAclTensor(input1HostData, selfShape1, input1DeviceAddr, aclDataType::ACL_FLOAT, input1); CHECK_RET(ret ACL_SUCCESS, return ret); // 创建input2 aclTensor ret CreateAclTensor(input2HostData, selfShape2, input2DeviceAddr, aclDataType::ACL_FLOAT, input2); CHECK_RET(ret ACL_SUCCESS, return ret); std::vectoraclTensor* tempInput{input1, input2}; aclTensorList* tensorListInput aclCreateTensorList(tempInput.data(), tempInput.size()); // 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aclnnForeachLogInplace第一段接口 ret aclnnForeachLogInplaceGetWorkspaceSize(tensorListInput, workspaceSize, executor); CHECK_RET(ret ACL_SUCCESS, LOG_PRINT(aclnnForeachLogInplaceGetWorkspaceSize 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aclnnForeachLogInplace第二段接口 ret aclnnForeachLogInplace(workspaceAddr, workspaceSize, executor, stream); CHECK_RET(ret ACL_SUCCESS, LOG_PRINT(aclnnForeachLogInplace 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侧内存上的结果复制至host侧需要根据具体API的接口定义修改 auto size GetShapeSize(selfShape1); std::vectorfloat self1Data(size, 0); ret aclrtMemcpy(self1Data.data(), self1Data.size() * sizeof(self1Data[0]), input1DeviceAddr, size * sizeof(self1Data[0]), ACL_MEMCPY_DEVICE_TO_HOST); for (int64_t i 0; i size; i) { LOG_PRINT(out1 result[%ld] is: %f\n, i, self1Data[i]); } size GetShapeSize(selfShape2); std::vectorfloat self2Data(size, 0); ret aclrtMemcpy(self2Data.data(), self2Data.size() * sizeof(self2Data[0]), input2DeviceAddr, size * sizeof(self2Data[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(out2 result[%ld] is: %f\n, i, self2Data[i]); } // 6. 释放aclTensor需要根据具体API的接口定义修改 aclDestroyTensorList(tensorListInput); // 7.释放device资源需要根据具体API的接口定义修改 aclrtFree(input1DeviceAddr); aclrtFree(input2DeviceAddr); if (workspaceSize 0) { aclrtFree(workspaceAddr); } aclrtDestroyStream(stream); aclrtResetDevice(deviceId); aclFinalize(); return 0; }【免费下载链接】ops-nn本项目是CANN提供的神经网络类计算算子库实现网络在NPU上加速计算。项目地址: https://gitcode.com/cann/ops-nn创作声明：本文部分内容由AI辅助生成（AIGC），仅供参考