30#ifndef ONNX_MODELRUNNER_H
31#define ONNX_MODELRUNNER_H
45 std::map<std::string, Agent *>
agents,
46 llvm::LLVMContext *
Ctx =
nullptr);
57 std::map<std::string, Agent *>
agents;
The MLModelRunner class is the main interface for interacting with the ML models.
Agent class to support ML/RL model inference via ONNX.
Agent is a wrapper around the ONNXModel class, interfaces with the Environment class to support ML/RL...
MLModelRunner - The main interface for interacting with the ML models.
ONNXModelRunner is the main user facing class that interfaces with the Environment and Agent classes ...
void * evaluateUntyped() override
Should be implemented by the derived class to call the model and get the result.
void setEnvironment(MLBridge::Environment *_env)
void requestExit() override
MLBridge::Environment * env
ONNXModelRunner(MLBridge::Environment *env, std::map< std::string, Agent * > agents, llvm::LLVMContext *Ctx=nullptr)
MLBridge::Environment * getEnvironment()
void addAgent(Agent *agent, std::string name)
std::map< std::string, Agent * > agents
void computeAction(Observation &obs)
Base Environment class to support ONNX based inference of RL models.
llvm::SmallVector< float, 300 > Observation