Fritz for Python

Faster mobile machine learning workflows

Posted Embed

Fritz for Python helps speed up your model development workflow by reducing barriers between model training and application development.

Reviews
Discussion
You need to become a Contributor to join the discussion.
Chris Kelly
Chris KellyHunter@chris_kelly90
Hey Hunters! Today we are releasing the Fritz Python Library to help speed up your model development workflow by reducing barriers between model training and application development. Using our Python library, you can easily convert and push models from your training scripts right to mobile devices for testing. Updating a model in a mobile app can be difficult and tedious, especially if you’re not a mobile developer. Everytime you want to try something new you need to export a model checkpoint, convert it to a mobile format, manually add it to the project in your IDE, and wait for the entire application to rebuild. The Fritz Python Library does all of this for you - instantly deploy new versions of a model your mobile app without ever leaving your Python workflow. We also provide integrations with popular training frameworks, like Keras, so the latest version of your model is always available in your app to test instantly. You can store metadata with each model so you never lose track of hyperparameters or experiments. See how it works by creating your own style transfer model using our Google Colab Style Transfer Notebook (https://colab.research.google.co...).