Coder Social home page Coder Social logo

Comments (4)

albertwujj avatar albertwujj commented on July 21, 2024

Could be why #16

from rainbow.

Kaixhin avatar Kaixhin commented on July 21, 2024

It should be handled by these conditions which make sure that samples near the index are discarded and resampled. Go ahead and work through the logic (I'm not 100% certain that this is bug-free, but I believe it's fine).

from rainbow.

albertwujj avatar albertwujj commented on July 21, 2024

Cool! On a first read it looks correct (I will certainly let you know if I discover its not).

I'm modifying your ReplayBuffer to allow parallel envs (still store all transitions in a 1D array, but step back num_envs indices rather than 1 index to get previous timestep). The code you linked is actually a place I missed updating, leading to a bug (getting None when looking for n-step transition states). Thank you so much for the help!

from rainbow.

albertwujj avatar albertwujj commented on July 21, 2024

Oops, I see there is a whole paper on 'distributed' replay: https://arxiv.org/pdf/1803.00933.pdf

from rainbow.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.