jorisvandenbossche commented on issue #25684: URL: https://github.com/apache/arrow/issues/25684#issuecomment-1620642472
With recent versions of both numpy and pyarrow (installed with conda-forge), I personally observe that pyarrow is faster than numpy on my laptop: ``` In [2]: %timeit -n 5 -r 5 df.multiply(df) 429 ms ± 21.1 ms per loop (mean ± std. dev. of 5 runs, 5 loops each) In [3]: %timeit -n 5 -r 5 pc.multiply(table[0],table[0]) 273 ms ± 7.45 ms per loop (mean ± std. dev. of 5 runs, 5 loops each) In [4]: np.__version__ Out[4]: '1.24.3' In [5]: pa.__version__ Out[5]: '12.0.0' ``` Anyway, given that there is nothing clearly actionable (apart from general performance improvements, which are welcome for all compute kernels), going to close this issue. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
