OCR 跨平台工程化onnxruntime gpu c++代码
网上关于onnxruntime运行在gpu环境下运行OCR模型的代码较少,经过查阅,其实就是在初始化模型的时候这样写就可以了
void DbNet::setNumThread(int numOfThread) {
numThread = numOfThread;
//===session options===
// Sets the number of threads used to parallelize the execution within nodes
// A value of 0 means ORT will pick a default
//sessionOptions.SetIntraOpNumThreads(numThread);
//set OMP_NUM_THREADS=16
// Sets the number of threads used to parallelize the execution of the graph (across nodes)
// If sequential execution is enabled this value is ignored
// A value of 0 means ORT will pick a default
sessionOptions.SetInterOpNumThreads(numThread);
// Sets graph optimization level
// ORT_DISABLE_ALL -> To disable all optimizations
// ORT_ENABLE_BASIC -> To enable basic optimizations (Such as redundant node removals)
// ORT_ENABLE_EXTENDED -> To enable extended optimizations (Includes level 1 + more complex optimizations like node fusions)
// ORT_ENABLE_ALL -> To Enable All possible opitmizations
sessionOptions.SetGraphOptimizationLevel(GraphOptimizationLevel::ORT_ENABLE_EXTENDED);
}
void DbNet::initModel(const std::string &pathStr) {
#ifdef _WIN32
std::wstring dbPath = strToWstr(pathStr);
session = new Ort::Session(env, dbPath.c_str(), sessionOptions);
#else
OrtSessionOptionsAppendExecutionProvider_CUDA(sessionOptions, 0);
session = new Ort::Session(env, pathStr.c_str(), sessionOptions);
#endif
getInputName(session, inputName);
getOutputName(session, outputName);
}
OrtSessionOptionsAppendExecutionProvider_CUDA(sessionOptions, 0);
这行代码就是将模型运行在gpu0上
sessionOptions.SetInterOpNumThreads(numThread);
这行代码是设置op线程数量
在crnn识别的时候,使用了batch作为预测的方法
所有代码的工程文件均可以在我的资料里面下载到
|