wolpertinger_ddpg icon indicating copy to clipboard operation
wolpertinger_ddpg copied to clipboard

Published 20 hours ago • verified publisher icon ChangyWen
→

Metadata

Wolpertinger Training with DDPG (Pytorch), Deep Reinforcement Learning in Large Discrete Action Spaces. Multi-GPU/Singer-GPU/CPU compatible.

Results 4 wolpertinger_ddpg issues
Sort by recently updated
recently updated
newest added

Question about Wolpertinger DDPG

I send email to you on 2022/07/22 Please check. Email is qq.com right ? Please contact with me thx.

Emissary112 avatar Emissary112

assert(npts >= num_neighbors) AssertionError

When I change the k_ration in agrs to generate multiple actions, AssertionError shows as: Traceback (most recent call last): File "/Users/xx/Downloads/DROO-master/mec/rlmodel/LDAS/main.py", line 211, in train(args.train_iter, agent, env, evaluate, File "/Users/xx/Downloads/DROO-master/mec/rlmodel/LDAS/main.py",...

Reeseee avatar Reeseee

action mask

Hi, If not all actions are valid, how can I apply action mask to filter invalid actions? Do you have any good idea to modify your code?

Xiong5Heng avatar Xiong5Heng

Update train_test.py

I believe that the episode_steps and episode_reward should be equal to zero after each episode finish

abdor95 avatar abdor95

← Metadata

64
Stars
16
Forks
Watchers

Owner

verified publisher icon ChangyWen

Metadata

Wolpertinger Training with DDPG (Pytorch), Deep Reinforcement Learning in Large Discrete Action Spaces. Multi-GPU/Singer-GPU/CPU compatible.

Back