|
Hey I am using your package as part of my work and must say it is amazing with the amount of speedup as compared to a naive python implementation. I had one doubt, in my Reinforcement Learning paradigm I have multiple transaction which I store by going individually in a loop. This unnecessary adds to the run time. Is there a way to add multiple transactions simultaneously to the Buffer in a vector operations form? |
Answered by
ymd-h
Nov 27, 2021
Replies: 1 comment
|
@pranshu93
In example code, 10 transitions are added with single method. If you still have any questions, please ask us freely. |
0 replies
Answer selected by
ymd-h
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
@pranshu93
Thank you for your positive feedback. I am very happy to hear that.
ReplayBuffer(and its sub-classes) can accept multiple transitions at once.In example code, 10 transitions are added with single method.
If you still have any questions, please ask us freely.