Skip to content

DCGAN

Directory actions

More options

Directory actions

More options

Latest commit

 

History

History

DCGAN

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
check
check
 
 
eager
eager
 
 
graph
graph
 
 
test
test
 
 
README.md
README.md
 
 
models.py
models.py
 
 
plot.py
plot.py
 
 
test_images.png
test_images.png
 
 
utils.py
utils.py
 
 

README.md

DCGAN

DCGAN is an unconditional image generation method (code in lazy mode), our code is inspired by TensorFlow Tutorial.

Please make sure your working directory is models/DCGAN/.

Train on mnist Dataset

Run Oneflow Training script

Eager Mode

bash eager/train.sh

Graph Mode

bash graph/train.sh

Inference

Run Oneflow Inference script

bash test/test_of_dcgan.sh

Check

Run Oneflow Check script

bash check/check.sh

Inference results

Eager Mode

test_images

Graph Mode

image_100