Transform ML models into native code with zero dependencies
1–10 of 27 posts
Re: Transform ML models into native code with zero dependencies
#2This is pretty cool idea! Would there be any possibility of trying to convert a model to javascript/node?
Haven't looked through the source but it looks like the generated code is a essentially the weights from a trained model transformed into a function for the target language.
Re: Transform ML models into native code with zero dependencies
#3Other similar projects:
https://github.com/nok/sklearn-porter Supports many scikit-learn models to Java/C/JavaScript/Go/Ruby, at least since 2016.
https://github.com/konstantint/SKompiler transpiles to Excel/SQL
https://github.com/jonnor/emlearn To C only, focus on microcontrollers/embedded devices. Includes feature extraction tools also. Disclaimer: I wrote it.
Re: Transform ML models into native code with zero dependencies
#4Other similar projects: https://github.com/nok/sklearn-porter Supports many scikit-learn models to Java/C/JavaScript/Go/Ruby, at least since 2016. https://github.com/konstantint/SKompiler transpiles to Excel/SQL https://github.com/jonnor/emlearn To C only, focus on microcontrollers/embedded devices. Includes feature extraction tools also. Disclaimer: I wrote it.
Thanks for sharing this! Looking at sklearn-porter now, hope I can contribute to the either the Ruby, Golang, or PHP library.
Re: Transform ML models into native code with zero dependencies
#5What are the advantages?
Re: Transform ML models into native code with zero dependencies
#6What are the advantages?
Trained ML models that run as native code with zero dependencies.
Re: Transform ML models into native code with zero dependencies
#7Re: Transform ML models into native code with zero dependencies
#8Similar: tfcompile AOT compiles TensorFlow models into native code using XLA (https://www.tensorflow.org/xla/tfcompile).
Re: Transform ML models into native code with zero dependencies
#9What are the advantages?
Trained ML models that run as native code with zero dependencies.
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.
Re: Transform ML models into native code with zero dependencies
#10Thanks for sharing. This is quite interesting as we’ve been mulling over library support for our platform.