El Bruno

Hi!

When a platform or technology begins to be popular, it often happens that Frameworks supporting this technology begin to appear as mushrooms in a wet forest in spring. JavaScript may be the best example of this.

Well, the same thing is happening in the world of artificial intelligence. The most common tools I found that specialists uses are Caffe2, CNTK, TensorFlow, and others. However, the interoperability or change between these tools is a problem with not an easy solution.

Well, 2 big players like Facebook and Microsoft, have agreed to help with the migration and interaction between these tools creating an interesting project ONNX.

ONNX is a community project created by Facebook and Microsoft. We believe there is a need for greater interoperability in the AI tools community. Many people are working on great tools, but developers are often locked in to one framework or ecosystem. ONNX is the first step in enabling more of these tools to work together by allowing them to share models. Our goal is to make it possible for developers to use the right combinations of tools for their project. We want everyone to be able to take AI from research to reality as quickly as possible without artificial friction from tool chains.

ONNX allows models to be trained in one framework and then transferred to another. The models currently compatible with Caffe2, CNTK, MXNet and PyTorch. In addition, there are connectors for many other frameworks.

Looking at the GitHub repository, it really impressed me that import and export scenarios from TensorFlow (Google) are included in the product roadmap.

Happy Coding!

Greetings @ Calgary

El Bruno

References

• Open Neural Network Exchange, Home
• Open Neural Network Exchange, GitHub
• Open Neural Network Exchange, Tutorials
• Caffe2
• CNTK
• TensorFlow