GithubHelp home page GithubHelp logo

testc6's Introduction

testC6

This is a test for learning Transfer Learning with the model of inception_v3.

The photo folder “flower_photos” (hidden) is a folder of test data which need to be downloaded from Google, and the checkpoint file inception_v3.ckpt should also be downloaded from Google.

The result just not good with a final accuracy rate below 0.3, so is there anybody who can help me in finding bugs causing this result?

2018/7/13:

I finally found out where the problem is. All the variates in the inception_v3.ckpt are defined under a scope, so that I should define the model under the scope too by using "with slim.arg_scope(inception_v3_arg_scope()):".

See:https://stackoverflow.com/questions/39357454/restore-checkpoint-in-tensorflow-tensor-name-not-found and the answer 2 in that.

testc6's People

Contributors

fishlikeapple avatar

Watchers

 avatar

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.