Skip to content
/ text-to-image-with-residual-attention-and-USE Public

text to image with the concept of residual attention and Universal sentence encoder

0 stars 0 forks Branches Tags Activity
Star
Notifications You must be signed in to change notification settings

labcontext/text-to-image-with-residual-attention-and-USE

BranchesTags

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

20 Commits

.idea

.idea

 
 

images

images

 
 

saved_model

saved_model

 
 

README.md

README.md

 
 

SpectralNorm.py

SpectralNorm.py

 
 

data_loader.py

data_loader.py

 
 

dcgan_from_keras.py

dcgan_from_keras.py

 
 

gen_caption_from_coco.py

gen_caption_from_coco.py

 
 

make_embedd.py

make_embedd.py

 
 

make_embedd.txt

make_embedd.txt

 
 

preprocess.py

preprocess.py

 
 

resat_gan.py

resat_gan.py

 
 

temp.py

temp.py

 
 

trial_gan.ipynb

trial_gan.ipynb

 
 

Repository files navigation

text-to-image-GAN-with-residual-attention-and-USE

코드를 돌리기 위한 준비물

  1. caption 과 image 쌍으로 이루어진 데이터셋
  2. 충분한 컴퓨팅파워가 보장된 환경

준비물 다운로드 링크 https://drive.google.com/drive/folders/1VZ7v28G-ekuYkCPnCxFjfyixSb4ypvrj?usp=sharing

코드 돌려보기

  1. trial_gan.ipynb 실행 (2. 미완성_코드)

시도했던 것과 시도하고 있는 것

  1. generator 내부 구조에 residual attention 적용하여 모델링해보기
  2. inferSent 또는 USE를 활용하여 sentence embedding 개선해보기
  3. 부가적인 모델 개선 수단 적용해보기
  4. 등등 다양한 구조 변화 및 모델 규제

About

text to image with the concept of residual attention and Universal sentence encoder

Resources

Readme
Activity
Custom properties

Stars

0 stars

Watchers

1 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

No packages published

Languages