Re: [sage-support] Straightforward calculation going wrong?
On Mon, 02 May 2011 at 04:28PM -0700, KvS wrote: > I am staring already for half an hour at the following. This piece of > code: > > reset() > print 'Loop 1:' > B=-2 > for A in range(-3,-1): > C=-A*B/(A+B) > print "A:",A,"B:",B,"C:",C,"A*B+C*(A+B)=",A*B+C*(A+B) > > print 'Loop 2:' > for A in range(-3,-1): > for B in range(-3,-1): > C=-A*B/(A+B) > print "A:",A,"B:",B,"C:",C,"A*B+C*(A+B)=",A*B+C*(A+B) > > should in both loops (obviously) always assign that value to C such > that A*B+C*(A+B) is 0. This indeed happens in the first loop, but not > in the second. Here is the output (Kubuntu 11.04, Sage v. 4.6.2): I think this is a preparsing issue. In loop 1, when you do B=-2, Sage preparses it so that B is a "Sage integer". In loop 2, the outputs from range() are Python integers. The difference is that Sage integers become rationals when divided, but Python integers do truncated division. Try changing loop 2 to: for B in [-3..-2]: or changing the assignment to C = -A * B / ZZ(A + B). Dan -- --- Dan Drake - http://mathsci.kaist.ac.kr/~drake --- signature.asc Description: Digital signature
Re: [sage-support] Straightforward calculation going wrong?
The "range" function is a Python one, and it returns Python ints.
Python ints have truncating division, so that 3/2 = 1, not 3/2. When
you type 3/2 at the Sage command, it's preparsed to be Sage Integers:
sage: 3/2
3/2
sage: preparse("3/2")
'Integer(3)/Integer(2)'
sage: int(3)/int(2)
1
sage: 3r/2r
1
(I don't know what the "r" is supposed to stand for but I always think
of it as "raw", i.e. Python, not Sage. I might even be right.) So in
your code your C values are wrong in the second loop. They work in
the first loop because "B=-2" makes it a Sage Integer, and so the
results of divisions can become Rationals like you expect.
print 'Loop 2:'
for A in range(-3,-1):
for B in range(-3,-1):
Cint = -A*B/(A+B)
CInt = (1)*-A*B/(A+B)
print A, B,
print 'Cint:', Cint, A*B+Cint*(A+B),
print 'CInt:', CInt, A*B+CInt*(A+B)
gives:
-3 -3 Cint: 1 3 CInt: 3/2 0
-3 -2 Cint: 1 1 CInt: 6/5 0
-2 -3 Cint: 1 1 CInt: 6/5 0
-2 -2 Cint: 1 0 CInt: 1 0
The multiplication by the Sage Integer 1 makes the CInt expression a
Sage one, and so it works. For these reasons, I tend to avoid using
"range" in Sage code entirely. There are several
alternatives. There's srange/xsrange=sxrange
sage: srange(-3, -1)
[-3, -2]
sage: sxrange(-3, -1)
sage: list(sxrange(-3, -1))
[-3, -2]
(The "s" stands for Sage):
Or you can be explicit and use IntegerRange, which gives a more
informative string:
sage: IntegerRange(-3, -1)
{-3, -2}
I like the "(a..b)" and "[a..b]" syntaxes myself:
sage: for A in (-3..-2): print A, type(A)
:
-3
-2
but note that it includes the right hand limit. I find this is more
useful in mathematical code than in pure programming, mostly because I
often find cases where I want (a..a) to include a, but your mileage
may vary.
Does that make sense?
Doug
--
To post to this group, send email to sage-support@googlegroups.com
To unsubscribe from this group, send email to
sage-support+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/sage-support
URL: http://www.sagemath.org
[sage-support] Straightforward calculation going wrong?
Dear all, I am staring already for half an hour at the following. This piece of code: reset() print 'Loop 1:' B=-2 for A in range(-3,-1): C=-A*B/(A+B) print "A:",A,"B:",B,"C:",C,"A*B+C*(A+B)=",A*B+C*(A+B) print 'Loop 2:' for A in range(-3,-1): for B in range(-3,-1): C=-A*B/(A+B) print "A:",A,"B:",B,"C:",C,"A*B+C*(A+B)=",A*B+C*(A+B) should in both loops (obviously) always assign that value to C such that A*B+C*(A+B) is 0. This indeed happens in the first loop, but not in the second. Here is the output (Kubuntu 11.04, Sage v. 4.6.2): Loop 1: A: -3 B: -2 C: 6/5 A*B+C*(A+B)= 0 A: -2 B: -2 C: 1 A*B+C*(A+B)= 0 Loop 2: A: -3 B: -3 C: 1 A*B+C*(A+B)= 3 A: -3 B: -2 C: 1 A*B+C*(A+B)= 1 A: -2 B: -3 C: 1 A*B+C*(A+B)= 1 A: -2 B: -2 C: 1 A*B+C*(A+B)= 0 What is happening here? I am sorry for obviously completely overlooking something trivial... Many thanks for your help, Kees -- To post to this group, send email to sage-support@googlegroups.com To unsubscribe from this group, send email to sage-support+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-support URL: http://www.sagemath.org
