numexpr
numexpr copied to clipboard
pydata
Fast numerical array expression evaluator for Python, NumPy, Pandas, PyTables and more
`expm1(1j*x)` has the same (low) accuracy for small complex arguments as `exp(1j*x) - 1`. For real arguments, the accuracy is as expected and on par with numpy: Code to reproduce:...
Seems like you can build against MKL by using `python setup.py install` still, but the new system designed around `pyproject.toml` and `setup.cfg` seems to ignore the effort to build MKL...
Is there a way to do ```python x[...] = y.T ``` using numexpr? From the answer given here https://stackoverflow.com/questions/67431966/how-to-avoid-huge-overhead-of-single-threaded-numpys-transpose I believe there might be something to gain using numexpr for...
This complies with the numpy main branch, but did not test with older versions of numpy (I suspect it wont work due to missing functions) The `np_2_compat.h` is designed to...
I'm currently testing numpy 2 on the the openSUSE python ecosystem. I notice the pandas test suite failing when numpy 2.0.0rc1 is installed instead of 1.26.4: ``` [ 681s] ===================================...
Could you please provide guidance as to citation preferences? Ideally, this repository would include a `CITATION.cff`, per [this GitHub blog post](https://github.blog/2021-08-19-enhanced-support-citations-github/).
Working on https://github.com/pandas-dev/pandas/issues/21374 I am Coming across unexpected complex128 upcasting. here is an example of numexpr upcasting complex64 to complex128 when python operation conserves the correct dtype. ```python >>> compl...
I suppose it's related to #479: ``` $ python -c 'import numexpr; numexpr.test()' -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-= Numexpr version: 2.10.1 NumPy version: 2.0.0 Python version: 3.12.4 (main, Jun 12 2024, 18:58:22) [GCC 14.1.1...
