GithubHelp home page GithubHelp logo

Comments (6)

L-M-Sherlock avatar L-M-Sherlock commented on August 26, 2024

Update the parameters once a month is enough. For long-term study, there is no need to reschedule all cards. Only reschedule all cards when you want to achieve your desired retention as soon as possible.

from fsrs4anki.

user1823 avatar user1823 commented on August 26, 2024
  • "Reschedule all cards" should be used only after updating the parameters or desired retention. However, you can use "reschedule cards reviewed in last x days" daily if you want to use load balancing. In this case, I recommend setting a low value of x (such as 1, 2 or 3) in the add-on config so that you don't unnecessarily reschedule cards multiple times.
  • You should update the parameters only after you have done at least 4000 reviews since the last optimization. For example, if you do 100 cards each day, re-optimize not earlier than 40 days after the last optimization. Otherwise, chances are high that you will get worse parameters after optimization.

For future questions, please use Anki Forums (https://forums.ankiweb.net/).

from fsrs4anki.

ratman-codes avatar ratman-codes commented on August 26, 2024

Thanks for the prompt answer. I'd like to ask one follow-up, why does it matter that you've done 4k reviews since last optimization? Doesn't it look at the entire histroy of reviews? curious how the optimization parameters could get "worse"

from fsrs4anki.

user1823 avatar user1823 commented on August 26, 2024

The optimizer can't guarantee that it finds the "best" set of parameters, though it can find one of the best ones. So, if you optimize again after a short time, there are chances that the new set of parameters is worse than the previous one.

If you re-optimize after doing a lot of reviews, the chances of getting better parameters increases. The 4000 reviews is an experimentally obtained threshold beyond which the likelihood of getting better parameters is high.

In addition, it seems that there is some systemic problem with the latest version of Anki that is increasing the chances of getting worse parameters on re-optimization. We are not yet sure what the problem is.

For more discussion, refer to the forum post: https://forums.ankiweb.net/t/measures-to-prevent-new-fsrs-parameters-from-being-worse-than-the-old-ones/38957

from fsrs4anki.

brishtibheja avatar brishtibheja commented on August 26, 2024

@user1823 I accidently and part curiously did optimize the parameters twice in one day (when I started using fsrs). what should I do now or if there is anything I can do to mitigate some of the harm I've done?

from fsrs4anki.

aleksejrs avatar aleksejrs commented on August 26, 2024

@brishtibheja The chances of the new parameters being worse are not because you optimized after a short time, but because the result is kind of fuzzy, so if there is no much change, it might end up a little below the average quality. Usually it's just a waste of time to optimize too often.

from fsrs4anki.

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.