I Don't Like NumPy

1. xarray

   1 14 4,193 9.4 Python

   N-D labeled arrays and datasets in Python

   

   If your arrays have more than two dimensions, please consider using Xarray [1], which adds dimension naming to NumPy arrays. Broadcasting and alignment then becomes automatic without needing to transpose, add dummy axes, or anything like that. I believe that alone solves most of the complaints in the article.

   Compared to NumPy, Xarray is a little thin in certain areas like linear algebra, but since it's very easy to drop back to NumPy from Xarray, what I've done in the past is add little helper functions for any specific NumPy stuff I need that isn't already included, so I only need to understand the NumPy version of the API well enough one time to write that helper function and its tests. (To be clear, though, the majority of NumPy ufuncs are supported out of the box.)

   I'll finish by saying, to contrast with the author, I don't dislike NumPy, but I do find its API and data model to be insufficient for truly multidimensional data. For me three dimensions is the threshold where using Xarray pays off.

   [1] https://xarray.dev

2. AppSignal

   www.appsignal.com sponsored

   AppSignal knows why the f*#k it crashed. Stop vibe-debugging. Every exception, every backtrace, grouped so you see patterns, not noise.

   

3. numpysane

   2 5 47 3.1 Python

   more-reasonable core functionality for numpy

   

   Hear hear! Some of these complaints have been resolved with numpysane: https://github.com/dkogan/numpysane/ . With numpysane and gnuplotlib, I now find numpy acceptable and use it heavily for everything. But yeah; without these it's unusable.

4. Numba

   3 130 11,144 9.9 Python

   NumPy aware dynamic Python compiler using LLVM

   

   Have you heard of JIT libraries like numba (https://github.com/numba/numba)? It doesn't work for all python code, but can be helpful for the type of function you gave as an example. There's no need to rewrite anything, just add a decorator to the function. I don't really know how performance compares to C, for example.

5. quickr

   4 1 185 9.2 R

   R to Fortran Transpiler

   

   Or, don't even write the fortran manually, just transpile the R function to fortran: https://github.com/t-kalinowski/quickr

6. array-api

   5 1 278 6.6 Python

   RFC document, tooling and other content related to the array API standard

   

7. Pytorch

   6 425 102,755 10.0 Python

   Tensors and Dynamic neural networks in Python with strong GPU acceleration

   

   I tried to do something similar with 'first-class' dimension objects in PyTorch https://github.com/pytorch/pytorch/blob/main/functorch/dim/R... .

8. nptyping

   7 3 638 0.0 Python

   💡 Type hints for Numpy and Pandas

   

   You could use third party library like https://github.com/ramonhagenaars/nptyping or https://github.com/beartype/beartype#numpy-arrays but it will not extend to the methode of Numpy.

9. Kargo

   akuity.io sponsored

   Stop Scripting Promotions. Start Shipping with Kargo. Kargo automates promotion across dev, staging, and prod with approval gates and verification. Open source, built by the team behind Argo CD. Download now.

   

10. beartype

    8 27 3,490 9.4 Python

    Unbearably fast near-real-time pure-Python runtime-static type-checker.

    

    You could use third party library like https://github.com/ramonhagenaars/nptyping or https://github.com/beartype/beartype#numpy-arrays but it will not extend to the methode of Numpy.

11. AxisArrays.jl

    9 2 207 3.1 Julia

    Performant arrays where each dimension can have a named axis with values

    

12. tsalib

    10 2 268 10.0 Python

    Tensor Shape Annotation Library (numpy, tensorflow, pytorch, ...)

    

13. tensor_annotations

    11 3 159 5.8 Python

    Discontinued Annotating tensor shapes using Python types

    

14. penzai

    12 6 1,894 6.3 Python

    A JAX research toolkit for building, editing, and visualizing neural networks.

    

15. mesh

    13 7 1,602 3.2 Python

    Discontinued Mesh TensorFlow: Model Parallelism Made Easier (by tensorflow)

    

16. torchdim

    14 3 321 2.4 Jupyter Notebook

    Discontinued Named tensors with first-class dimensions for PyTorch

    

17. SaaSHub

    www.saashub.com sponsored

    SaaSHub - Software Alternatives and Reviews. SaaSHub helps you find the best software and product alternatives

    

Suggest a related project