search

Kaixhin / Rainbow

Rainbow: Combining Improvements in Deep Reinforcement Learning
MIT License
1.59k stars 284 forks source link

Infinite loop in ReplayMemory._get_sample_from_segment #60

Closed luisenp closed 4 years ago

luisenp commented 4 years ago

I'm testing a variant of this code using the vizdoom environment and in some cases I see an infinite loop here https://github.com/Kaixhin/Rainbow/blob/master/memory.py#L103. I'm still not familiar with the code so not sure how to go about debugging this. Any pointers would be appreciated. (Note that I haven't changed any of the replay memory code).

Thanks.

Kaixhin commented 4 years ago

Duplicate of #41 . As noted there, it's an inherent problem with using PER, which can be resolved by setting hyperparameters appropriately for the domain.