On Saturday, 18 April 2015 03:34:57 UTC+1, Ian wrote: > On Fri, Apr 17, 2015 at 7:19 PM, Paddy <paddy3...@..l.com> wrote: > > Having just seen Raymond's talk on Beyond PEP-8 here: > > https://www.youtube.com/watch?v=wf-BqAjZb8M, it reminded me of my own > > recent post where I am soliciting opinions from non-newbies on the relative > > Pythonicity of different versions of a routine that has non-simple array > > manipulations. > > > > The blog post: > > http://paddy3118.blogspot.co.uk/2015/04/pythonic-matrix-manipulation.html > > > > The first, (and original), code sample: > > > > def cholesky(A): > > L = [[0.0] * len(A) for _ in range(len(A))] > > for i in range(len(A)): > > for j in range(i+1): > > s = sum(L[i][k] * L[j][k] for k in range(j)) > > L[i][j] = sqrt(A[i][i] - s) if (i == j) else \ > > (1.0 / L[j][j] * (A[i][j] - s)) > > return L > > > > > > The second equivalent code sample: > > > > def cholesky2(A): > > L = [[0.0] * len(A) for _ in range(len(A))] > > for i, (Ai, Li) in enumerate(zip(A, L)): > > for j, Lj in enumerate(L[:i+1]): > > s = sum(Li[k] * Lj[k] for k in range(j)) > > Li[j] = sqrt(Ai[i] - s) if (i == j) else \ > > (1.0 / Lj[j] * (Ai[j] - s)) > > return L > > > > > > The third: > > > > def cholesky3(A): > > L = [[0.0] * len(A) for _ in range(len(A))] > > for i, (Ai, Li) in enumerate(zip(A, L)): > > for j, Lj in enumerate(L[:i]): > > #s = fsum(Li[k] * Lj[k] for k in range(j)) > > s = fsum(Lik * Ljk for Lik, Ljk in zip(Li, Lj[:j])) > > Li[j] = (1.0 / Lj[j] * (Ai[j] - s)) > > s = fsum(Lik * Lik for Lik in Li[:i]) > > Li[i] = sqrt(Ai[i] - s) > > return L > > > > My blog post gives a little more explanation, but I have yet to receive any > > comments on relative Pythonicity. > > I prefer the first version. You're dealing with mathematical formulas > involving matrices here, so subscripting seems appropriate, and > enumerating out rows and columns just feels weird to me. > > That said, I also prefer how the third version pulls the last column > of each row out of the inner loop instead of using a verbose > conditional expression that you already know will be false for every > column except the last one. Do that in the first version, and I think > you've got it.
But shouldn't the maths transcend the slight change in representation? A programmer in the J language might have a conceptually neater representation of the same thing due to its grounding in arrays (maybe) and for a J representation it would become J-thonic. In Python, it is usual to iterate over collections and also to use enumerate where we must have indices. Could it be that there is a also a strong pull in the direction of using indices because that is what is predominantly given in the way matrix maths is likely to be expressed mathematically? A case of "TeX likes indices so we should too"? -- https://mail.python.org/mailman/listinfo/python-list
