I Use This!
Activity Not Available
Analyzed over 2 years ago. based on code collected over 2 years ago.

Project Summary

What it isThe numexpr package evaluates multiple-operator array expressions many times faster than NumPy can. It accepts the expression as a string, analyzes it, rewrites it more efficiently, and compiles it to faster Python code on the fly. It's the next best thing to writing the expression in C and compiling it with a specialized just-in-time (JIT) compiler, i.e. it does not require a compiler at runtime.

Also, numexpr has support for the Intel VML (Vector Math Library) -- integrated in Intel MKL (Math Kernel Library) --, allowing nice speed-ups when computing transcendental functions (like trigonometrical, exponentials...) on top of Intel-compatible platforms. This support also allows to use multiple cores in your computations.

Why It WorksThere are two extremes to array expression evaluation. Each binary operation can run separately over the array elements and return a temporary array. This is what NumPy does: 2*a + 3*b uses three temporary arrays as large as a or b. This strategy wastes memory (a problem if the arrays are large). It is also not a good use of CPU cache memory because the results of 2*a and 3*b will not be in cache for the final addition if the arrays are large.

The other extreme is to loop over each element:

for i in xrange(len(a)):
c[i] = 2*a[i] + 3*b[i]This conserves memory and is good for the cache, but on each iteration Python must check the type of each operand and select the correct routine for each operation. All but the first such checks are wasted, as the input arrays are not changing.

numexpr uses an in-between approach. Arrays are handled in chunks (the first pass uses 256 elements). As Python code, it looks something like this:

for i in xrange(0, len(a), 256):
r0 = a[i:i+256]
r1 = b[i:i+256]
multiply(r0, 2, r2)
multiply(r1, 3, r3)
add(r2, r3, r2)
c[i:i+256] = r2The 3-argument form of add() stores the result in the third argument, instead of allocating a new array. This achieves a good balance between cache and branch prediction. The virtual machine is written entirely in C, which makes it faster than the Python above.

For more info about numexpr, read the Numexpr's Overview written by the original author (David M. Cooke).

Examples of UseUsing it is simple:

>>> import numpy as np
>>> import numexpr as ne

>>> a = np.arange(1e6) # Choose large arrays for high performance
>>> b = np.arange(1e6)

>>> ne.evaluate("a + 1") # a simple expression
array([ 1.00000000e+00, 2.00000000e+00, 3.00000000e+00, ...,
9.99998000e+05, 9.99999000e+05, 1.00000000e+06])

>>> ne.evaluate('a*b-4.1*a > 2.5*b') # a more complex one
array([False, False, False, ..., True, True, True], dtype=bool)and fast... :-)

>>> timeit a**2 + b**2 + 2*a*b
10 loops, best of 3: 33.3 ms per loop
>>> timeit ne.evaluate("a**2 + b**2 + 2*a*b")
100 loops, best of 3: 7.96 ms per loop # 4.2x faster than NumPy


numpy pytables python scipy


MIT License

Commercial Use



Private Use



Hold Liable


Include Copyright

Include License

These details are provided for information only. No information here is legal advice and should not be used as such.

All Licenses

This Project has No vulnerabilities Reported Against it

Did You Know...

  • ...
    Black Duck offers a free trial so you can discover if there are open source vulnerabilities in your code
  • ...
    learn about Open Hub updates and features on the Open Hub blog
  • ...
    use of OSS increased in 65% of companies in 2016
  • ...
    data presented on the Open Hub is available through our API



30 Day Summary

Mar 17 2016 — Apr 16 2016

12 Month Summary

Apr 16 2015 — Apr 16 2016
  • 97 Commits
    Up + 78 (410%) from previous 12 months
  • 19 Contributors
    Up + 13 (216%) from previous 12 months