Skip to content

ConvDRAW: “Towards Conceptual Compression” (NIPS 2016) with TensorFlow.

License

Notifications You must be signed in to change notification settings

geosada/ConvDRAW

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

17 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

ConvDRAW

TensorFlow Implementation of DeepMind’s Convolutional DRAW: “Towards Conceptual Compression” (NIPS 2016) on the SVHN generation task.

The DRAW process proceeds in 8 steps from the leftmost column to the second column from the left and the rightmost column is original images.

 

This code was based on Eric Jang's DRAW.

Usage

python main.py

ToDo

There seems to be something bug in tensor multiplication in write attention function therefore it turns off now.

Useful Resources

About

ConvDRAW: “Towards Conceptual Compression” (NIPS 2016) with TensorFlow.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages