Coder Social home page Coder Social logo

crf-chain2-tiers's People

Contributors

kawu avatar tindzk avatar

Stargazers

avatar

Watchers

avatar avatar avatar

Forkers

tindzk

crf-chain2-tiers's Issues

Compilation issues

Due to changes in Vector, there are some problems during compilation:

Resolving dependencies...
Configuring crf-chain2-tiers-0.2.1...
Building crf-chain2-tiers-0.2.1...
Failed to install crf-chain2-tiers-0.2.1
Build log ( /Users/tim/.cabal/logs/crf-chain2-tiers-0.2.1.log ):
Configuring crf-chain2-tiers-0.2.1...
Building crf-chain2-tiers-0.2.1...
Preprocessing library crf-chain2-tiers-0.2.1...
[ 4 of 10] Compiling Data.CRF.Chain2.Tiers.Dataset.Internal ( src/Data/CRF/Chain2/Tiers/Dataset/Internal.hs, dist/build/Data/CRF/Chain2/Tiers/Dataset/Internal.o )

src/Data/CRF/Chain2/Tiers/Dataset/Internal.hs:68:16:
    Could not coerce from ‘Data.Vector.Primitive.Vector
                             Int32’ to ‘U.Vector Ob’
      because ‘Data.Vector.Primitive.Vector Int32’ and ‘U.Vector
                                                          Ob’ are different types.
      arising from the coercion of the method ‘G.basicLength’ from type
                   ‘U.Vector Int32 -> Int’ to type ‘U.Vector Ob -> Int’
    Possible fix:
      use a standalone 'deriving instance' declaration,
        so you can specify the instance context yourself
    When deriving the instance for (G.Vector U.Vector Ob)

[...]

DAG: print the actual accurracy when training

Currently, when computing the accuracy (f = ...) of the model, we assume that the model will determine the most probably label for each edge in the DAG. This does not make sense for edges which are not selected by the model.

We should determine, for each edge in the DAG:

  • Is it actually an element of the most probable path?
    • If not, Nothing.
    • If yes, Just the most probable label assigned by the model to this edge.

Also, we could consider (or make it possible and optional) printing the likelihood of the evaluation set.

Improve pruning

While discarding the model features, the codec could be trimmed as well by removing atoms which are no longer in use. It would also probably require renumbering of the atoms.

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.