2 ms·
Yes. I mean it's fine, but I don't see any reason to use it instead of Python, and lots of reasons not to. But I'm not a mathematician by training. I do quite
by nl 6y ago
Yes.
I mean it's fine, but I don't see any reason to use it instead of Python, and lots of reasons not to. But I'm not a mathematician by training.
I do quite like RStudio though, and I do see places where that is useful. So maybe MALBAB fits somewhere in between - less stats than R, less programmign that Python.
- langitbiru 6y agoWhat do you think of Julia?
- nl 6y ago> Julia The Haskell of Machine Learning. https://www.jwz.org/doc/worse-is-better.html https://www.jwz.org/doc/worse-is-better.html
- mkl 6y agoI am a mathematician by training (+CS), I've worked in Matlab quite a bit, I've taught hundreds of students in it (having no say in language, but we're about to switch to Python), and I probably see even more reasons not to use it than you do. (I haven't done ML in it, but I'd be astounded if it wasn't terrible compared to Python and its ecosystem.) I recommend avoiding Matlab for every use case unless they've got you trapped with a huge existing code base or reliance on a proprietary toolbox.
- alexilliamson 6y agoIMO whenever RStudio decides to support Jupyter notebooks, it's game over for everyone else. It's such a great piece of software for data analysis and I hope they continue to go broader than the just the R language.