Code Monkey home page Code Monkey logo

Comments (3)

dragen1860 avatar dragen1860 commented on July 21, 2024

  1. index of losses_q is NOT tasks index, but the update step index. sry for wrong comments.

  2. sum of loss_q is accumulated at every inner update, hence just need to average it.
    losses_q[-1] is the last step's loss.

3.yes, you are right. For only single step's update setting, the code loss_q = F.cross_entropy(logits_q, y_qry[i]) should be moved out of torch.no_grad().

from maml-pytorch.

yaox12 avatar yaox12 commented on July 21, 2024

At the end of each task i, you append loss_q to list losess_q

MAML-Pytorch/meta.py

Lines 130 to 131 in fc20b31

# 4. record last step's loss for task i
losses_q.append(loss_q)

The accumulated last step's loss should be losses_q[self.update_step] instead of losses_q[-1], because the length of losses_q is update_step + 1 + task_num in the end.
In fact, I think the above two lines are redundant and useless.

from maml-pytorch.

dragen1860 avatar dragen1860 commented on July 21, 2024

Yes, it's a bug!
Thanks for your very helpful insight.
Remove line 130 & 131 !

from maml-pytorch.

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.