ADEO Dev Summit

As a developer, you may follow the hype and add machine learning capabilities to your apps; but do you know how to do that?

Tooling, mathematical concepts, Python, equations, backpropagation, linear regression, Neural Networks, Tensorflow... Where is the starting point?
Do you have to lose yourself in a brand new world to use Machine Learning in your code?

No! Nowadays, it exists a plethora of pre-trained models; and the good news is that it is getting easy to use them. Hopefully, Go can help us to run them at scale.

In this talk, I introduce you to the basic concepts of neural networks and their relation to the "software 2.0".
Then I introduce ONNX, an independent way to serialize a machine learning model.
Hence, I present my project onnx-go, which is an interface that allows to read (and executes) neural networks exported from dominant python framework into your regular go code.