Giter VIP home page Giter VIP logo

vision-bert-ccl's Introduction

vision-bert-CCL

Prepare dataset and models

Pretrain Dataset:

conceptual caption dataset, you can download the precomputed features through link (the files are being uploaded)

then unzip your downloaded file, obtain the fold cc, and move it to the fold ./dataset

Finetune Dataset:

MS coco dataset, you can download the precomputed features through this link

Alternatively, you can obtain it in command line by

wget https://www.dropbox.com/s/4g0x49h302xj7j4/coco.tar.gz?dl=0

then unzip your downloaded file, obtain the fold coco, and move it to the fold ./dataset

BERT Model

you can download the pretrained bert model through this link

then move your downloaded .bin file to ./bert fold

Pretrained Vision-Bert Model

you can download the pretrained bert model through this link

then move your downloaded .bin file to ./models fold

Run Scripts

Pretrain the vision-bert

python train.py --task pretrain --mlm --cm

Finetune the pretrained model on the retrieval task:

python train.py --task retrieval --resume models/6layer_mlm_pretrain.pth.tar

To do

Ploy Vision-Text Encoder

vision-bert-ccl's People

Contributors

ntuyt avatar

Stargazers

 avatar  avatar  avatar

Watchers

 avatar  avatar  avatar

vision-bert-ccl's Issues

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    ๐Ÿ–– 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. ๐Ÿ“Š๐Ÿ“ˆ๐ŸŽ‰

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 โค๏ธ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.