The unitree_il_lerobot open-source project is a modification of the LeRobot open-source training framework, enabling the training and testing of data collected using the dual-arm dexterous hands of Unitree's G1 robot.
关键参数与规格
关键参数与规格
参数
值
语言
Python
类别
Model Development
子类别
Reinforcement Learning
关键词
RL
星标数
779
分叉数
113
贡献者
4
创建日期
2018-12-09
更新日期
2026-08-31
1 天星标数
0
7 天星标数
0
排名
13,909
上次排名
13,793
排名变化
-116
7 天星标排名
14,180
上次 7 天星标排名
14,199
7 天星标排名变化
19
总星标排名
13,639
上次总星标排名
13,623
总星标排名变化
-16
国家
South Korea
README 摘录
Let's do Inverse RL Introduction This repository contains PyTorch (v0. 1) implementations of **Inverse Reinforcement Learning (IRL)** algorithms. Apprenticeship Learning via Inverse Reinforcement Learning [2] Maximum Entropy Inverse Reinforcement Learning [4] Generative Adversarial Imitation Learning [5] Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow [6] We have implemented and trained the agents with the IRL algorithms using the following environments. OpenAI GYM Mountain car Mujoco Hopper For reference, reviews of below papers related to
IRL (in Korean) are located in Let's do Inverse RL Guide. , "Algorithms for Inverse Reinforcement Learning", ICML 2000. , "Apprenticeship Learning via Inverse Reinforcement Learning", ICML 2004. , "Maximum Entropy Inverse Reinforcement Learning", AAAI 2008. , "Generative Adversarial Imitation Learning", NIPS 2016. , "Variational Discriminator Bottleneck. Improving Imitation Learning, Inverse RL, and GANs by Constraining Information Flow", ICLR 2019. Table of Contents Let's do Inverse RL Introduction Table of Contents Mountain car Mujoco Hopper Reference Implementation team members Mountain car We have implemented , using
Q-learning as RL step in environment. Expert's demonstrations Navigate to in lets-do-irl/mountaincar/app/expert_demo or lets-do-irl/mountaincar/maxent/expert_demo. Shape of expert's demonstrations is (20, 130, 3); (number of demonstrations, length of demonstrations, states and actions of demonstrations) If you make demonstrations, Navigate to in lets-do-irl/mountaincar/app/expert_demo or lets-do-irl/mountaincar/maxent/expert_demo. Train & Test APP Navigate to lets-do-irl/mountaincar/app folder. **Train
** the agent wtih without rendering. If you want to test , **Test** the agent with the saved model in folder. MaxEnt Navigate to lets-do-irl/mountaincar/maxent folder. **Train** the agent wtih without rendering. If you want to test , **Test** the agent with the saved model in folder. Trained Agent We have trained the agents with two different IRL algortihms using environment. Installation Mac OS (in Korean) Ubuntu 2. Expert's demonstrations Navigate to in lets-do-irl/mujoco/gail/expert
_demo or lets-do-irl/mujoco/vail/expert_demo. Shape of expert's demonstrations is (50000, 14); (number of demonstrations, states and actions of demonstrations) We used demonstrations that get scores between about 2200 and 2600 on average. If you want to make demonstrations, Navigate to in lets-do-irl/mojoco/ppo folder. Also, you can see detailed implementation story (in Korean) of PPO in PG Travel implementation story. Train & Test GAIL Navigate to lets-do-irl/mujoco/gail folder. **Train
** the agent wtih without rendering. If you want to **Continue training** from the saved checkpoint, Note that file should be in the folder. If you want to test , **Test** the agent with the saved model in the folder. VAIL Navigate to lets-do-irl/mujoco/vail folder. **Train** the agent wtih without rendering. If you want to **Continue training** from the saved checkpoint, Note that file should be in the folder. If you want to test , **Test** the agent with
the saved model in the folder. Tensorboard Note that the results of trainings are automatically saved in folder. TensorboardX is
