Are you sure you want to delete this task? Once this task is deleted, it cannot be recovered.
Bo Zhou d33f30025c | 4 years ago | |
---|---|---|
.. | ||
.benchmark | 5 years ago | |
README.md | 4 years ago | |
actor.py | 4 years ago | |
atari_agent.py | 4 years ago | |
atari_model.py | 4 years ago | |
impala_config.py | 5 years ago | |
learner.py | 4 years ago | |
run_actors.sh | 5 years ago | |
train.py | 5 years ago |
Based on PARL, the IMPALA algorithm of deep reinforcement learning is reproduced, and the same level of indicators of the paper is reproduced in the classic Atari game.
Please see here to know more about Atari games.
Result with one learner (in a P40 GPU) and 32 actors (in 32 CPUs).
PongNoFrameskip-v4: mean_episode_rewards can reach 18-19 score in about 7~10 minutes.
Results of other games in an hour.
python train.py
for i in $(seq 1 32); do
python actor.py &
done;
wait
You can change training settings (e.g. env_name
, server_ip
) in impala_config.py
.
Training result will be saved in log_dir/train/result.csv
.
PARL 是一个高性能、灵活的强化学习框架
Python C++ JavaScript Shell Markdown other
Dear OpenI User
Thank you for your continuous support to the Openl Qizhi Community AI Collaboration Platform. In order to protect your usage rights and ensure network security, we updated the Openl Qizhi Community AI Collaboration Platform Usage Agreement in January 2024. The updated agreement specifies that users are prohibited from using intranet penetration tools. After you click "Agree and continue", you can continue to use our services. Thank you for your cooperation and understanding.
For more agreement content, please refer to the《Openl Qizhi Community AI Collaboration Platform Usage Agreement》