Matt Feinstein wrote:
> On Mon, 16 May 2005 12:03:24 -0600, Steven Bethard
> <[EMAIL PROTECTED]> wrote:
>
>>Can you give an example of what you really want to do? Probably there
>>are numarray functions that you can use. In general, you'll do better
>>applying a sequence of numarray functions than operating element-wise on
>>an array and converting it from a list back to an array...
>
> Well, for example, suppose I want to modify the elements of the matrix
> in some fashion. However, I'm not entirely sure how I want to do it.
> As a strawman, I generate a function with a Boolean test in it that
> multiplies by one factor if the matrix element is in an interval and
> by a different factor if it is outside the interval
As Steven Bethard suggests, using the ufuncs provided to express the
function in a vectorial way is always the best option *if* it's possible.
Otherwise, you may want to look at Scipy's vectorize() function. I don't
think the numarray port is quite working yet, so you may have to use
Numeric. You will still have the same function call overhead since you
will be calling the Python function for each element, but the loops will
be in C.
Type: classobj
String Form: scipy_base.function_base.vectorize
Namespace: Interactive
File:
/Library/Frameworks/Python.framework/Versions/2.4/lib/python2.4/site-packages/scipy_base/function_base.py
Docstring:
vectorize(somefunction) Generalized Function class.
Description:
Define a vectorized function which takes nested sequence objects or
numerix arrays as inputs and returns a numerix array as output,
evaluating the function over successive tuples of the input arrays like
the python map function except it uses the broadcasting rules of numerix
Python.
Input:
somefunction -- a Python function or method
Example:
def myfunc(a,b):
if a > b:
return a-b
else:
return a+b
vfunc = vectorize(myfunc)
>>> vfunc([1,2,3,4],2)
array([3,4,1,2])
--
Robert Kern
[EMAIL PROTECTED]
"In the fields of hell where the grass grows high
Are the graves of dreams allowed to die."
-- Richard Harter
--
http://mail.python.org/mailman/listinfo/python-list
