CANN/ops-nn自适应平均池化3D反向计算

aclnnAdaptiveAvgPool3dBackward【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 查看源码产品是否支持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aclnnAdaptiveAvgPool3d 的反向计算。函数原型每个算子分为两段式接口必须先调用“aclnnAdaptiveAvgPool3dBackwardGetWorkspaceSize”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“aclnnAdaptiveAvgPool3dBackward”接口执行计算。aclnnStatus aclnnAdaptiveAvgPool3dBackwardGetWorkspaceSize( const aclTensor *gradOutput, const aclTensor *self, aclTensor *out, uint64_t *workspaceSize, aclOpExecutor **executor)aclnnStatus aclnnAdaptiveAvgPool3dBackward( void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, aclrtStream stream)aclnnAdaptiveAvgPool3dBackwardGetWorkspaceSize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TensorgradOutput输入当前节点的梯度。数据类型与self一致NC维度和self保持一致。FLOAT32、FLOAT16、BFLOAT16NCDHW、ND4-5√self输入叶子节点。-FLOAT32、FLOAT16、BFLOAT16NCDHW、ND4-5√out输出对应了输入叶子节点的梯度。-FLOAT32、FLOAT16、BFLOAT16NCDHW、ND4-5√workspaceSize输出返回需要在Device侧申请的workspace大小。-----executor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gradOutput、self或out是空指针。ACLNN_ERR_PARAM_INVALID161002gradOutput和self的数据类型和数据格式不在支持的范围之内。gradOutput、self和out数据类型不一致。gradOutput、self和out的维数不等于4或5。gradOutput、self和out的shape不匹配。gradOutput或self的shape的某一维不大于0。gradOutput和self的数据格式不一致。NC维度不一致。Ascend 950PR/Ascend 950DT: gradOutputself的shapeN轴取值可以为0aclnnAdaptiveAvgPool3dBackward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aclnnAdaptiveAvgPool3dBackwardGetWorkspaceSize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aclnnAdaptiveAvgPool3dBackward默认非确定性实现支持通过aclrtCtxSetSysParamOpt开启确定性。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 cstdio #include iostream #include vector #include math.h #include acl/acl.h #include aclnnop/aclnn_adaptive_avg_pool3d_backward.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::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. device/stream初始化参考acl API手册 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 std::vectorint64_t yGradShape {2, 2, 1, 1, 2}; std::vectorint64_t xShape {2, 2, 1, 1, 4}; std::vectorint64_t xGradShape {2, 2, 1, 1, 4}; void* yGradDeviceAddr nullptr; void* xDeviceAddr nullptr; void* xGradDeviceAddr nullptr; aclTensor* yGrad nullptr; aclTensor* x nullptr; aclTensor* xGrad nullptr; std::vectorfloat yGradHostData {1, 2, 3, 4, 5, 6, 7, 8}; std::vectorfloat xHostData(GetShapeSize(xShape), 1); std::vectorfloat xGradHostData(16, 0); // 创建yGrad aclTensor ret CreateAclTensor(yGradHostData, yGradShape, yGradDeviceAddr, aclDataType::ACL_FLOAT, yGrad); CHECK_RET(ret ACL_SUCCESS, return ret); // 创建x aclTensor ret CreateAclTensor(xHostData, xShape, xDeviceAddr, aclDataType::ACL_FLOAT, x); CHECK_RET(ret ACL_SUCCESS, return ret); // 创建xGrad aclTensor ret CreateAclTensor(xGradHostData, xGradShape, xGradDeviceAddr, aclDataType::ACL_FLOAT, xGrad); CHECK_RET(ret ACL_SUCCESS, return ret); // 3. 调用CANN算子库API uint64_t workspaceSize 0; aclOpExecutor* executor; // 调用aclnnAdaptiveAvgPool3dBackward第一段接口 ret aclnnAdaptiveAvgPool3dBackwardGetWorkspaceSize(yGrad, x, xGrad, workspaceSize, executor); CHECK_RET(ret ACL_SUCCESS, LOG_PRINT(aclnnAdaptiveAvgPool3dBackwardGetWorkspaceSize 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aclnnAdaptiveAvgPool3dBackward二段接口 ret aclnnAdaptiveAvgPool3dBackward(workspaceAddr, workspaceSize, executor, stream); CHECK_RET(ret ACL_SUCCESS, LOG_PRINT(aclnnAdaptiveAvgPool3dBackward failed. ERROR: %d\n, ret); return ret); // 4. 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 auto size GetShapeSize(xGradShape); std::vectorfloat resultData(size, 0); ret aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), xGradDeviceAddr, 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 aclDestroyTensor(yGrad); aclDestroyTensor(x); aclDestroyTensor(xGrad); // 7. 释放Device资源需要根据具体API的接口定义修改 aclrtFree(yGradDeviceAddr); aclrtFree(xDeviceAddr); aclrtFree(xGradDeviceAddr); 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