在Ubuntu上安裝d4rl
阿新 • • 發佈:2022-04-20
在安裝好mujoco,mujoco_py的基礎上,安裝d4rl,D4RL的github:https://github.com/rail-berkeley/d4rl 有介紹安裝方法,但直接安裝會報各種錯誤。
1.先安裝兩個庫檔案,避免報錯 pip install absl-py pip install matplotlib
2.安裝 dm_control pip install dm_control ,如果不先安裝dm_control,會報錯:
3. 克隆D4RL倉庫:
git clone https://github.com/rail-berkeley/d4rl.git
4.找到d4rl目錄下的setup.py檔案,註釋mujoco_py, dm_control, mjrl:
注意一定要註釋掉mjrl,否則會報錯:
5. 安裝:
pip install -e .
6.安裝mjrl : pip install git+https://github.com/aravindr93/mjrl@master#egg=mjrl
7. 測試:
import gym
import d4rl # Import required to register environments
# Create the environment
env = gym.make('maze2d-umaze-v1')
# d4rl abides by the OpenAI gym interface
env.reset()
env.step(env.action_space.sample())
# Each task is associated with a dataset
# dataset contains observations, actions, rewards, terminals, and infos
dataset = env.get_dataset()
print(dataset['observations']) # An N x dim_observation Numpy array of observations
# Alternatively, use d4rl.qlearning_dataset which
# also adds next_observations.
dataset = d4rl.qlearning_dataset(env)
8. 最終效果:
參考文獻:
1. 離線強化學習(Offline RL)系列2: (環境篇)D4RL資料集簡介、安裝及錯誤解決 ( https://zhuanlan.zhihu.com/p/489475047)
2. 安裝Mujoco、Mujoco_py、D4RL、gym、d3rlpy以及Pycharm遠端連線伺服器問題 (https://zhuanlan.zhihu.com/p/434073300)