Skip to content

Flask Deployment Code For Image Classifiers in Seconds

License

Notifications You must be signed in to change notification settings

ashishbairwa/Katana

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

29 Commits
 
 
 
 
 
 

Repository files navigation

License: MIT


Flask Deployment Code For Image Classifiers Built with Pytorch in Seconds


Katana is a little script which helps you to generate a boilerplate (More than a boilerplate) to directly deploy your model in the clouds. It works pretty well with render. All you need to do is to answer some quick questions and Katana will generate templates, commons.py and everything for you just in seconds.

How to use it ?

It's pretty easy. Download the katana.py, place it in a folder and answer the questions accordingly. Here is a video of demonstration https://www.youtube.com/watch?v=VqeD8lRhl2o&feature=youtu.be

What things i need in advance ?

You need to place your checkpoint file generated by your model. Place it in the same folder and upload everything on the git.

Screenshots

ss1

asd

ss3

ss4

res


What's Next ?

  • Video illustration and examples
  • More theme variants - (adding a random gradient generator)
  • Support to nn.Sequential writing mode
  • Support to Keras and FastAI

PR and Issues are warmly welcome! Cheers!

About

Flask Deployment Code For Image Classifiers in Seconds

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages