使用tensorflow::Session是要釋放的，否則循環起來，崩潰到懷疑人生。

方法一：
使用Close()

	tensorflow::Session * session;tensorflow::Status status; tensorflow::GraphDef graphdef; //Define a graph for the current modelstatus = tensorflow::NewSession(tensorflow::SessionOptions(), &session); // Create a new session Sessiontensorflow::Status status_load = tensorflow::ReadBinaryProto(tensorflow::Env::Default(), model_path, &graphdef); // read graph model from pb filetensorflow::Status status_create = session->Create(graphdef); // Import the model into the session Session	status_run = session->Run(...);status = session->Close();delete session;

方法二：
使用智能指針

	tensorflow::SessionOptions options;tensorflow::Status status; tensorflow::GraphDef graphdef; //Define a graph for the current modelstd::unique_ptr<tensorflow::Session> session = std::unique_ptr<tensorflow::Session>(tensorflow::NewSession(options)); // Create a new session Sessiontensorflow::Status status_load = tensorflow::ReadBinaryProto(tensorflow::Env::Default(), model_path, &graphdef); // read graph model from pb filetensorflow::Status status_create = session->Create(graphdef); // Import the model into the session Session	status_run = session->Run(...);

總結

以上是生活随笔為你收集整理的Tensorflow::Session 释放内存的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。