Transform ML models into native code with zero dependencies

Re:Transform ML models into native code with zero dependencies by mahomedalid on Tuesday March 05, 2019 @12:18AM (#19308275)

What are the advantages?

[ Parent | on HN ]

Re:Transform ML models into native code with zero dependencies by leetbulb on Tuesday March 05, 2019 @12:48AM (#19308388)

Trained ML models that run as native code with zero dependencies.

[ Parent | on HN ]

Re:Transform ML models into native code with zero dependencies by danieldk on Tuesday March 05, 2019 @01:36AM (#19308543)

Note though that e.g. liblinear models are trivial to load and apply yourself - most classifiers just compute the matrix-vector product between the weight matrix and an instance vector and take the class with the highest activation. That route has the benefit that you do not hardcode a model, but can easily load new models.

Not to criticize this project. This looks nice and has many useful applications.

[ Parent | on HN ]

Re:Transform ML models into native code with zero dependencies by TuringNYC on Tuesday March 05, 2019 @01:14AM (#19308470)

Heck, you could even put it into a lambda function on a lambda service and deploy in like 5 min! That is a lot of value.

[ Parent | on HN ]

Re:Transform ML models into native code with zero dependencies by darius_morawiec on Tuesday March 05, 2019 @06:49AM (#19309897)

Yes, you got to the heart of it. *

* owner and main contributor of sklearn-porter

[ Parent | on HN ]

Re:Transform ML models into native code with zero dependencies by rob_fulwell on Tuesday March 05, 2019 @12:24PM (#19312194)

How does the runtime performance compare? (mem/cpu)

[ Parent | on HN ]