Code Monkey home page Code Monkey logo

Comments (2)

matt-gardner avatar matt-gardner commented on June 29, 2024 1

Hmm, interesting. Thanks for finding this! I dug into it a bit and discovered that it's due to the relu activation in the feedforward layer. The initialization happens to be such that the activations going into the relu are all negative, and so the gradient gets zeroed out. This is only an issue because the network is so small, and the particulars of how the random seed gets used must have changed since we first wrote the test, so now we get an initialization that consistently fails.

I fixed this by switching the relu activation to a sigmoid activation, and now the tests pass.

from allennlp-as-a-library-example.

murphp15 avatar murphp15 commented on June 29, 2024

Thanks!

from allennlp-as-a-library-example.

Related Issues (11)

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.