Code Monkey home page Code Monkey logo

vision-language's Introduction

GQA: Visual Reasoning in the Real World

Data structure

├── Question Number
    ├── Annotations
    |   ├── answer
    |   ├── full Answer
    |   └── question
    │   
    ├── answer
    ├── entailed
    ├── equivalent
    ├── fullAnswer
    ├── groups
    ├── imageId
    ├── isBalanced
    ├── question
    ├── semantic
    ├── semanticStr
    └── types
        ├── detailed
        ├── semantic
        └── structural
  • answer
  • imageId
  • question

Network Architecture

캡처

Image-Question Aggregator

캡처2

Requirements

  • tensorflow-gpu==1.13.1
  • numpy==1.16.2
  • tensorflow-hub==0.4.0
  • python==3.7.3
  • cv2==4.0.0
  • tqdm==4.31.1

vision-language's People

Contributors

leaderj1001 avatar plemeri avatar

Stargazers

avatar avatar avatar

Watchers

avatar avatar

Forkers

hxdaze

vision-language's Issues

[PPT Contents]

  • GQA 설명 (2 장)
  • 네트워크 그림, 원래 하려고 했던 것에서 어떻게 바뀌어서 현재 사용되고 있는지? (3 장)
  • 진행상황 (1-2장)
    • 현재 학습 진행중이고,
    • 향후에 어떻게 할 예정이다 + CVPR Workshop 제출 예정

[Todo]

  • ",", "?"는 어떻게 처리하는게 좋은지? -> 바꾼다면 pkl도 다시 만들어야함.
  • ELMo, training 할 것인지, freezing할 것인지?
  • Epochs
  • max_len 조절 -> max_len: 29
  • batch iterator 시간 줄이기
  • reporting(loss, acc, tensorboard)
  • model save, load
  • augmentation

[Experiments]

태훈: token_weights 사용한 것 / 안한 것
명준: ELMo fine tuning 함.

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.