Kaixhin / PlaNet

Deep Planning Network: Control from pixels by latent planning with learned dynamics

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Random Chunk Sample Vs Random idx sample

idigitopia opened this issue · comments

commented

why do we sample a chunk of index list rather than just sampling random index in experience replay?

This is just code for sampling random indices, retrieving a certain sequence length around that index, parallelised over a batch.