search

openai / random-network-distillation

Code for the paper "Exploration by Random Network Distillation"
https://openai.com/blog/reinforcement-learning-with-prediction-based-rewards/
881 stars 159 forks source link

how to test? #12

Closed wangrui1996 closed 5 years ago

wangrui1996 commented 5 years ago

I can train,but i don't know how to test the general file and show test result by videos

wangrui1996 commented 5 years ago

I solved it by myself,I edit ./vec_env.py : in the def worker (remote ...) --208 function,add env.render() following ob, reward, done, info = ... fininally result show is: if cmd == 'step': ob, reward, done, info = env.step(data) env.render() if done: ob = env.reset() remote.send((ob, reward, done, info))