numexpr

numexpr copied to clipboard

add an example of using strings (literals) on Python3

2

_From [[email protected]](https://code.google.com/u/101943334880212128972/) on June 05, 2013 20:42:28_ discovered thru PyTables 3.00, issue here: https://github.com/PyTables/PyTables/issues/265 In Python 3.3 In [13]: import numpy as np In [14]: np.**version** Out[14]: '1.7.1' In [15]:...

FrancescAlted

_From [[email protected]](https://code.google.com/u/102653066256529587359/) on August 09, 2012 16:39:36_ What steps will reproduce the problem? a,b,c,d = True, False, "C", "D" numexpr.evaluate("where(a>b,c,d)") What is the expected output? What do you see instead?...

FrancescAlted

Can evalute expersions with record arrays

1

_From [[email protected]](https://code.google.com/u/114036687774939598344/) on May 31, 2012 13:51:33_ What steps will reproduce the problem? 1. create a record array img = numpy.zeros([10], dtype=[('r',float),('g',float),('b',float)]) img['r'] = numpy.arange(10) 2. call numexpr numexpr.evaluate("img['r'] \*...

FrancescAlted

_From [[email protected]](https://code.google.com/u/106920573304176801777/) on April 11, 2012 11:20:44_ What steps will reproduce the problem? sum and prod produce different results from numpy.sum and numpy.prod when reducing a dimension with a length...

FrancescAlted

support accumulate

2

_From [[email protected]](https://code.google.com/u/110941570796194873664/) on April 11, 2012 03:37:36_ I often process vectors of categorical information, and need to transform them (often as an intermediate step) into probabilities of being in a...

FrancescAlted

-sum(a) does not work

3

_From [[email protected]](https://code.google.com/u/111502234986314066506/) on March 19, 2012 15:03:51_ > > import numexpr as ne > > > > > ne.evaluate('-sum(1)') > > > ne.evaluate('-1_sum(1)') > > > ne.evaluate('sum(1)_(-1)') All of these...

FrancescAlted

float constant changes to integer in a**-1.0

2

_From [[email protected]](https://code.google.com/u/107312151144929335969/) on January 30, 2012 16:37:04_ This issue is related to issue 59 . evaluate('a**-1.0') differs from numpy's a**-1.0 More generally, observe: print 'NumPy:', np.**version**, 'NumExpr:', ne.**version** a =...

FrancescAlted

_From [[email protected]](https://code.google.com/u/116605672876360875462/) on January 10, 2012 12:28:31_ n/t _Original issue: http://code.google.com/p/numexpr/issues/detail?id=68_

FrancescAlted

Problem with "and" (also, __nonzero__ should not be defined)

5

_From [[email protected]](https://code.google.com/u/117706089323896263204/) on December 12, 2009 04:23:42_ > > > a = numpy.array([1, 2, 1, 2, 1], dtype=numpy.int8) # or numpy.int32 > > > b = numpy.array([1, 2, 3, 4,...

FrancescAlted

get_optimization() is called twice

1

_From [[email protected]](https://code.google.com/u/101083734699402587936/) on April 14, 2008 14:26:04_ If you apply the next patch to numexpr sources: {{{ --- numexpr/expressions.py ( revision 14 ) +++ numexpr/expressions.py (working copy) @@ -180,6 +192,7...

FrancescAlted