Switch lightgbm to random forest in modeling_test (#1442)

Author: linmagit

src/self_driving/model_server/model_server_manager.cpp

@@ -87,6 +87,12 @@ void ModelServerManager::StartModelServer(const std::string &model_path) {
   }
 
   // Fork success
+  // TODO(lin): The LightGBM (Python) library has a known hanging issue when multithreading and using forking in
+  //  Linux at the same time:
+  //  https://lightgbm.readthedocs.io/en/latest/FAQ
+  //  .html#lightgbm-hangs-when-multithreading-openmp-and-using-forking-in-linux-at-the-same-time
+  //  So this may cause a subsequent command that issues a model training using LightGBM under multi-threading to
+  //  hang (not always). Don't have a good solution right now.
   if (py_pid_ > 0) {
     // Parent Process Routine
     MODEL_SERVER_LOG_INFO("Model Server Process running at : {}", py_pid_);

test/self_driving_e2e/modeling_test.cpp

@@ -79,8 +79,8 @@ TEST_F(ModelServerTest, PipelineTest) {
   std::string msg = "ModelServerTest";
   ms_manager->PrintMessage(msg);
 
-  // Perform a training of the opunit models with {lr, gbm} as training methods.
-  std::vector<std::string> methods{"lr", "gbm"};
+  // Perform a training of the opunit models with {lr, rf} as training methods.
+  std::vector<std::string> methods{"lr", "rf"};
   std::string save_path = "model_server_test.pickle";
 
   ModelServerFuture<std::string> future;