PKU-EPIC/UniDexGrasp2

Issues

• Why the success rate of policies trained using a single object is still only 60%？

  #10 opened a month ago by Sjey-Lyn
  1

• Evaluation code problem，我只想使用训练好的策略进行评价，我在脚本文件中增加了test，和model_dir，为什么程序还是从零开始训练呢？

  #8 opened 3 months ago by RexDrac
  1

• Point Cloud 𝑷𝒕=0是同时包含手和物体吗？

  #7 opened 3 months ago by guoqingMeng
  1

• 数据集中使用的'object_euler_xy'和 'object_init_z'是怎么生成的

  #6 opened 3 months ago by guoqingMeng
  4

• About the sensors in the Allegro Hand

  #5 opened 3 months ago by xsaxasni
  0

• error gym

  #4 opened 5 months ago by yiqiaoqingyuyu
  4

• A Question About Dagger Value Algorithm

  #2 opened 9 months ago by yangyichu
  2

• [Bug] DAggerValue, ppo_buffer adds next_obs instead of current_obs

  #3 opened 9 months ago by Phimos
  0

• The code

  #1 opened a year ago by kaijieshi7
  1