[Python-ideas] FW: Inline Functions - idea

[Amber Yust]

In which case you could simply do e.g.:

    result = [[dot_product(i, j) for i, el in enumerate(row)] for j, row in
enumerate(matrix)]

On Mon Feb 10 2014 at 11:57:20 AM, Alex Rodrigues <lemiant at hotmail.com>
wrote:

> I have to agree about matrix_multiply, it's not that clean. But try
> writing it out without inline and you'll find that it is quite a bit
> longer. If we were aiming for more readability I suppose we could do this
>
> def matrix_multiply(matrix, matrix2):
>     def dot_product(i, j):
>         sum = 0
>         for x in range(len(matrix2)):
>             sum += matrix[i][x] * matrix2[x][j]
>         return sum
>
> result = copy.copy(matrix)
> element_wise(result[i][j] = dot_product(i, j))
> return result
>
> Which is still superior to the best syntax currently available. So being
> concerned about the readablility of the function, while legitimate, is not
> actually a knock on the syntax just on the way I wrote it.
>
> > ________________________________
> >> From: amber.yust at gmail.com
> >> Date: Mon, 10 Feb 2014 19:20:41 +0000
> >> Subject: Re: [Python-ideas] Inline Functions - idea
> >> To: lemiant at hotmail.com; python-ideas at python.org
> >>
> >> I find your example code far less readable than just writing out the
> >> loops in the appropriate functions, especially for matrix_multiply.
> >>
> >> On Mon Feb 10 2014 at 10:51:35 AM, Alex Rodrigues
> >> <lemiant at hotmail.com<mailto:lemiant at hotmail.com>> wrote:
> >> Over the last few days I've been considering a lot of what we talked
> >> about regarding Python scoping and inline functions. Today I found a
> >> further use case for inline functions (one which is a fair bit harder
> >> to replicate using standard techniques). I was working on writing a
> >> basic linear algebra calculator for school and one of the annoying
> >> things about doing it is that every time you want to do an element-wise
> >> operation on the matrix (which is a list of lists) you have to nest
> >> what is essentially one line of code inside a pair of for loops.
> >> Below I've done three basic linear algebra functions which are written
> >> as if there was a builtin "inline()" which transformed a standard
> >> function into an inline function and also assuming that python had more
> >> powerful lambdas. I feel that these represent a good use case, because
> >> the inline function serves to make the code both shorter and clearer.
> >>
> >>
> >> import copy
> >>
> >> @inline
> >> def element_wise(func):
> >> func = inline(func)
> >> for i, row in enumerate(matrix):
> >> result.append([0]*len(row))
> >> for j, el in enumerate(row):
> >> func()
> >>
> >> def is_symmetric(matrix):
> >> element_wise(lambda: if matrix[i][j] != matrix[j][i]: return False)
> >> return True
> >>
> >> def matrix_add(matrix, matrix2):
> >> result = copy.deepcopy(matrix)
> >> element_wise(lambda: result[i][j] = matrix[i][j] + matrix2[i][j])
> >> return result
> >>
> >> def matrix_multiply(matrix, matrix2):
> >> result = copy.deepcopy(matrix)
> >> element_wise(lambda: result[i][j] = sum([a*b for a,b in
> >> zip(matrix[i], [r[j] for r in matrix2])]))
> >> return result
> >> _______________________________________________
> >> Python-ideas mailing list
> >> Python-ideas at python.org<mailto:Python-ideas at python.org>
> >> https://mail.python.org/mailman/listinfo/python-ideas
> >> Code of Conduct: http://python.org/psf/codeofconduct/
> _______________________________________________
> Python-ideas mailing list
> Python-ideas at python.org
> https://mail.python.org/mailman/listinfo/python-ideas
> Code of Conduct: http://python.org/psf/codeofconduct/
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-ideas/attachments/20140210/bdecf151/attachment-0001.html>