dummy-gan-evaluation

App for evaluation images produced by GAN just by choosing which one is real

Config

Look at the .env.sample in server and client directories for all available configurations.

Server

yarn install

DEV

yarn start:dev

PROD

yarn build

Client

yarn install

DEV

yarn dev

PROD

yarn build && yarn start

Images

All images are placed inside public directory on the server. Real images are placed inside real subdirectory, fake inside fake.

Example tree (2 classes):

public
--- real
------  0
----------  img1r.png
------  1
----------  img1r.jpg
--- fake
------  0
----------  img1f.jpg
------  1
----------  img1f.png

Image's names does not matter. But class name directories should start from 0. There also no need to have exactly the same amount of images per class. Remember to create directory logs at server top-level.

License

MIT, Petro Liashchynskyi