PyTorch: Where we are headed and why it looks a lot like Julia (but not exactly)

Re:PyTorch: Where we are headed and why it looks a lot like ... by edgyquant on Friday November 26, 2021 @07:21PM (#29355794)

No these languages are the drivers of the tech revolution. PHP was widely adopted because it allowed people to rapidly build personal websites and then those people went on to build companies with it. Python is popular in ML because of its syntax being close to pseudo-code and allowing people who aren’t programmers to interact with all the old math libraries you used to have write C and Fortran to use. JavaScript is popular because it made the web more interactive and thus a ton of people who made websites learned it as their first programming language when they were young.

Python and PHP are so big because of the languages themselves and their implementation, Js is a bit different in that regard I’ll admit.

[ Parent | on HN ]

Re:PyTorch: Where we are headed and why it looks a lot like ... by forrestthewoods on Friday November 26, 2021 @07:44PM (#29355946)

It's true that the tech revolution would not have happened if all code had to be written in assembly. And it's true the machine learning revolution would not have happened if all code had to be written in C.

Is Python better than assembly and C/C++ for ML? Absolutely. Is Python good? I don't think so. Other people might use that term, I do not. I think Python is a bad language that would be designed very differently it was built today. And we're stuck with its decisions because the inertia to change them is monumental.

It's not really the language's fault. As an industry we've learned a lot in the past 30 years! It would be a travesty if we hadn't made language design process! Unfortunately we haven't figured out how to effectively apply lessons to encumbered ecosystems. Case in point: the transition from Python 2 to Python 3 was an absolute disaster. And it only made tiny, incremental improvements!

[ Parent | on HN ]