6 ms·

Pre-release of Python 3.11 sees a 10%-60% performance gain over Python 3.10

• anakaine 4y ago
  I'd like to know how this translates to other frequently used modules that are potentially computationally heavy, if at all. - Numpy. - Pandas. - Scikit. - Dask. - Flask / Django. Others?
  • pjmlp 4y ago
    Some of those are bindings to C and C++ libraries, so hardly any change.
  • IshKebab 4y ago
    Well the heavy lifting in the numerical libraries you listed is done in C so it won't make any difference at all.
• ralgozino 4y ago
  The enhanced error location in tracebacks is the real deal IMHO: https://docs.python.org/3.11/whatsnew/3.11.html#enhanced-error-locations-in-tracebacks https://docs.python.org/3.11/whatsnew/3.11.html#enhanced-err...
• zikohh 4y ago
  There is a lot of talk about how 3.11 is a good performance improvement over 3.10 and especially in comparison to other python minor version updates. Is there anyway to quantify other version performance increases i.e between 3.8 and 3.9 and 3.9 to 3.10 etc? So it's a nice comparison