Hey everyone! Just wanted to share something exciting for those of you working across multiple ML frameworks.
Ivy is a Python package that allows you to seamlessly convert ML models and code between frameworks like PyTorch, TensorFlow, JAX, and NumPy. With Ivy, you can take a model you’ve built in PyTorch and easily bring it over to TensorFlow without needing to rewrite everything. Great for experimenting, collaborating, or deploying across different setups!
On top of that, we’ve just partnered with Kornia, so now Kornia can also be used in TensorFlow, JAX, and NumPy. You can check it out in the latest Kornia release (v0.7.4) with the new methods:
kornia.to_tensorflow()kornia.to_jax()kornia.to_numpy()
It’s all powered by Ivy’s transpiler to make switching frameworks seamless. Give it a try and let us know what you think!
- Install Ivy:
pip install ivy - More info: Ivy on GitHub
- Ivy Demos: Demos
Happy experimenting!