Neo

Ok! thanks your reply! First, Let me try to restart epoch when memory is full. because the pre-trained model have get some skills, so every episode will hold the same...

Yeah, it's the PongNoFrameskip-v4, I use the default parameters, because of I use python==2.7 and torch==1.0.0, I only modify the function random.choices() in dqn_agent.py. And I have train the code...

Em, Ok , Thanks！Let me try the CartPole-v0 and LunarLander-v2 environment firstly, and then try others. thanks for your answer :)

Hello. Have you solve this problem?

have you installed pr2_description package??

hello, I have uploaded the parameters of my model, you can find it in the following folder: https://github.com/nizhihao/Collaborative-Pushing-Grasping/tree/master/myur_ws/src/drl_push_grasp/logs/collabrative-pushing-grasping/models

hello, do you want to run the reward_net? (train.py?) the error _OSError: [Errno 2] No such file or directory: '/home/turtlebot/new_drl_datasets(more_like_real)_before20201019/'_ happen because of the executable code, you can delete and...

hello, you can try the following solution: Notice: If compile error : dh_hand_driver/hand_state.h: No such file or directory please compile it again Or try : **catkin_make -j1**