Search results

Discounted reward calculation in PPO.py breaks when trajectory reaches max_ep_len in train.py

Status: Open.

#73 In nikhilbarhate99/PPO-PyTorch;

· CaoAnda opened on Nov 5, 2025

Environment setting about python version, gym and roboschool

Status: Open.

#70 In nikhilbarhate99/PPO-PyTorch;

· vickychen928 opened on Jun 5, 2024

为什么新的版本在k epochs更新时不重新计算advantages？

Status: Open.

#68 In nikhilbarhate99/PPO-PyTorch;

· 31CFDC30 opened on May 25, 2024

(Solved) No env.reset() at the end of each training epoch.

Status: Open.

#67 In nikhilbarhate99/PPO-PyTorch;

· slDeng1003 opened on Mar 11, 2024

the version problem about gym and roboschool

Status: Open.

#66 In nikhilbarhate99/PPO-PyTorch;

· ShunZuo-AI opened on Feb 8, 2024

policy_old完全看不出作用

Status: Open.

#65 In nikhilbarhate99/PPO-PyTorch;

· haduoken opened on Jan 25, 2024

ValueError: expected sequence of length 8 at dim 1 (got 0)

Status: Open.

#64 In nikhilbarhate99/PPO-PyTorch;

· kavinwkp opened on Nov 11, 2023

question

Status: Open.

#63 In nikhilbarhate99/PPO-PyTorch;

· yinshuangshuang621671 opened on Jun 27, 2023

optimize the existing Chinese generation model

Status: Open.

#62 In nikhilbarhate99/PPO-PyTorch;

· ARES3366 opened on Apr 25, 2023

Continuous action space should use Independent Normal instead of MultivariateNormal

Status: Open.

#60 In nikhilbarhate99/PPO-PyTorch;

· imathg opened on Nov 23, 2022

Test results are not good

Status: Open.

#59 In nikhilbarhate99/PPO-PyTorch;

· 295885025 opened on Oct 13, 2022

Would a shared network work ?

Status: Open.

#58 In nikhilbarhate99/PPO-PyTorch;

· Miguel-s-Amaral opened on Sep 11, 2022