Some Lesser Known Machine Learning Libraries
blog.paralleldots.com
Some Lesser Known Machine Learning Libraries
1–10 of 54 posts
Re: Some Lesser Known Machine Learning Libraries
#2If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
Re: Some Lesser Known Machine Learning Libraries
#3If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
I feel at this point this is not a sensible thing to do any more unfortunately. I totally get the impulse though. For instance, there is still nothing great on the JVM for deep learning with symbolic differentiation (deeplearning4j does not have this, correct me if this has changed).
On the other hand, I realize that between writing native interfaces, symbolic differentiation (e.g. writing a port of autograd), network optimisers, custom layers, parameter servers, multi-GPU scheduling and so forth, I'd spend years before getting to do what I wanted to implement in the first place.
Re: Some Lesser Known Machine Learning Libraries
#4If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
Probably a better thing to do would be to download one of the many excellent open source libraries and explore how they work. You can then even contribute back. Many open source projects would really benefit from new users contributing to documentation for a start.
Re: Some Lesser Known Machine Learning Libraries
#5If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
Probably a better thing to do would be to download one of the many excellent open source libraries and explore how they work. You can then even contribute back. Many open source projects would really benefit from new users contributing to documentation for a start.
Could you name a few such open source projects ?
Re: Some Lesser Known Machine Learning Libraries
#6If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
to code a machine learning library from scratch, you must first invent the universe
Re: Some Lesser Known Machine Learning Libraries
#7If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
I guess, it purely depends on what you hope to achieve. If you're going to spend a few months learning how ML works, sure you'll benefit immensely. But if you're planning to apply in some field/area, writing your own library and making sure it's better than others, well, it's not going to be easy or fast..
Re: Some Lesser Known Machine Learning Libraries
#8Earlier quoted context omitted.
Probably a better thing to do would be to download one of the many excellent open source libraries and explore how they work. You can then even contribute back. Many open source projects would really benefit from new users contributing to documentation for a start.
Could you name a few such open source projects ?
scikit-learn is a production ready library that has some very well commented and easy to read source code.
https://github.com/eriklindernoren/ML-From-Scratch is an easy to understand understand implementation of some the basic ML algorithms built from first principles and aims for readability over performance.
Re: Some Lesser Known Machine Learning Libraries
#9If I was going to try machine learning I would code up from scratch rather than relying on black boxes.
Please don't spread bad advice. There is a reason we teach Python and not ia64 Assembly to beginners.
Re: Some Lesser Known Machine Learning Libraries
#10My first reaction to this, "there must be a reason that they are not known", I click the link and "Error establishing a database connection" :D