用来研究机器人多任务学习和终身持续学习的知识迁移问题(有关对象或空间关系的陈述性知识和有关动作或行为的程序性知识)
提供:
相同任务内部的迁移
- LIBERO-Spatial
- LIBERO-Object
- LIBERO-Goal
不同任务间的迁移
- LIBERO-100:LIBERO-90用来训练,LIBERO-10用来验证
python==3.8.13
torch==1.11.0+cu113 torchvision==0.12.0+cu113 torchaudio==0.11.0
pip install robosuite
- 下载源码
pip install -e .
pip install bddl
在指定文件夹下创建datasets文件夹
运行导包命令查看是否安装成功
from libero.libero import benchmark
from libero.libero.envs import OffScreenRenderEnv
print('successful')
警告:
[robosuite WARNING] No private macro file found! (__init__.py:7)
[robosuite WARNING] It is recommended to use a private macro file (__init__.py:8)
[robosuite WARNING] To setup, run: python /home/jiangyvhang/anaconda3/envs/maniskill2/lib/python3.8/site-packages/robosuite/scripts/setup_macros.py (__init__.py:9)
按照提示运行指定python文件:
再次运行测试是否安装成功
from libero.libero import benchmark
from libero.libero.envs import OffScreenRenderEnv
print('successful')