aclnnEmbeddingRenorm
支持的产品型号
- 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“aclnnEmbeddingRenormGetWorkspaceSize”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“aclnnEmbeddingRenor”接口执行计算。
aclnnStatus aclnnEmbeddingRenormGetWorkspaceSize(aclTensor *selfRef, const aclTensor *indices, double maxNorm, double normType, uint64_t *workspaceSize, aclOpExecutor **executor)
aclnnStatus aclnnEmbeddingRenorm(void *workspace, uint64_t workspaceSize, aclOpExecutor *executor, const aclrtStream stream)
功能描述
算子功能:根据给定的maxNorm和normType返回输入tensor在制定indices下的修正结果。
计算公式:向量的范数计算公式如下,其中n为normType指定的范数值:
针对计算出的范数大于maxNorm的场景,需要做归一化处理,对indices指定的0维元素乘以系数:
aclnnEmbeddingRenormGetWorkspaceSize
参数说明:
- selfRef(aclTensor*, 计算输入/输出):待进行renorm计算的入参,dim必须为2。Device侧的aclTensor,数据类型支持bfloat16(仅Atlas A2训练系列产品/Atlas 800I A2推理产品支持)、float16、float32类型,数据格式支持ND,支持非连续的Tensor。
- indices(aclTensor*, 计算输入):selfRef中第0维上待进行renorm计算的索引。Device侧的aclTensor,数据类型支持int32、int64类型,数据格式支持ND,支持非连续的Tensor。indices中的索引数据不支持越界。
- maxNorm(double, 计算输入):指定范数的最大值,超出此值需要对embedding的结果进行归一化处理,仅支持double类型。
- normType(double, 计算输入):指定L_P范数的类型,仅支持double类型。
- 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。
161001(ACLNN_ERR_PARAM_NULLPTR):1. 传入的selfRef、indices是空指针。 161002(ACLNN_ERR_PARAM_INVALID):1. selfRef、indices、maxNorm、normType的数据类型和数据格式不在支持的范围之内。 2. selfRef的dim不为2、indices的dim超出8维。
aclnnEmbeddingRenorm
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aclnnEmbeddingRenorm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_embedding_renorm.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侧引擎
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() {
int32_t deviceId = 0;
aclrtStream stream;
auto ret = Init(deviceId, &stream);
CHECK_RET(ret == 0, LOG_PRINT("Init acl failed. ERROR: %d\n", ret); return ret);
std::vector<int64_t> selfShape = {4, 2};
std::vector<int64_t> indicesShape = {4, 2};
void* selfDeviceAddr = nullptr;
void* indicesDeviceAddr = nullptr;
aclTensor* self = nullptr;
aclTensor* indices = nullptr;
std::vector<float> selfHostData = {0, 1, 2, 3, 4, 5, 6, 7};
std::vector<int> indicesHostData = {1, 1, 1, 1, 0, 0, 0, 0};
float normType = 1.0f;
float maxNorm = 2.0f;
ret = CreateAclTensor(selfHostData, selfShape, &selfDeviceAddr, aclDataType::ACL_FLOAT, &self);
CHECK_RET(ret == ACL_SUCCESS, return ret);
ret = CreateAclTensor(indicesHostData, indicesShape, &indicesDeviceAddr, aclDataType::ACL_INT32, &indices);
CHECK_RET(ret == ACL_SUCCESS, return ret);
uint64_t workspaceSize = 0;
aclOpExecutor* executor;
ret = aclnnEmbeddingRenormGetWorkspaceSize(self, indices, maxNorm, normType, &workspaceSize, &executor);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEmbeddingRenormGetWorkspaceSize failed. ERROR: %d\n", ret); return ret);
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);
}
ret = aclnnEmbeddingRenorm(workspaceAddr, workspaceSize, executor, stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclnnEmbeddingRenorm failed. ERROR: %d\n", ret); return ret);
ret = aclrtSynchronizeStream(stream);
CHECK_RET(ret == ACL_SUCCESS, LOG_PRINT("aclrtSynchronizeStream failed. ERROR: %d\n", ret); return ret);
auto size = GetShapeSize(selfShape);
std::vector<float> resultData(size, 0);
ret = aclrtMemcpy(resultData.data(), resultData.size() * sizeof(resultData[0]), selfDeviceAddr, 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("result[%ld] is: %f\n", i, resultData[i]);
}
aclDestroyTensor(self);
aclDestroyTensor(indices);
aclrtFree(selfDeviceAddr);
aclrtFree(indicesDeviceAddr);
if (workspaceSize > 0) {
aclrtFree(workspaceAddr);
}
aclrtDestroyStream(stream);
aclrtResetDevice(deviceId);
aclFinalize();
return 0;
}