[sage-support] Re: plot3d and expression evaluation
I think I've got a vague idea what happens here... The same thing also happens when I define the max() function myself: sage: def my_max(x,y): sage: if(x>y): return x sage: else: return y sage: fermi2(x,y,d,L) = 1 - 1/( exp( ( my_max(abs(x),abs(y))-L) /d) + 1) sage: fermi2 (x, y, d, L) |--> -1/(e^(-(L - abs(y))/d) + 1) + 1 This looks a bit like max(x,y) gets passed the the objects 'abs(x)' and 'abs(y)' and decides, for whatever reason, that 'abs(x)' is not greater than 'abs(y)', so the else-clause gets executed and we are left with abs(y). Does this make sense to you? (I'm just thinking aloud, I got no idea how sage actually does it) Most likely the internal "max" function is defined in rather the same way (it just returns abs(x) instead of abs(y). ) The bottom line seems to be that it is a very dangerous game to mix symbolic expressions and python functions. Is there some way to control when (and how) Sage evaluates expressions and invokes python functions? Something like saying "first evaluate abs(x) and abs(y) and then invoke the python function max() with the result of the evaluation"? On Mar 12, 5:49 pm, Harald Schilly wrote: > Ah, I got it, the max function is evaluated during the definition of > the function, look: > > sage: fermi(x,y,d,L) = 1 - 1/( exp( ( max(abs(x),abs(y))-L) /d) + 1) > sage: fermi > (x, y, d, L) |--> -1/(e^(-(L - abs(x))/d) + 1) + 1 > > especially: > sage: max(abs(x),abs(y)) > abs(x) > > There is no y any more! > > H -- 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] Re: plot3d and expression evaluation
On Mar 12, 5:47 pm, Harald Schilly wrote: > I'm not sure but there might be a bug or problem evaluating the > expression. Anyways, going the "pure" python way works: > > sage: def fermi(x,y,d,L): return 1 - 1/( exp( ( max(abs(x),abs(y))- > L) /d) + 1) > > sage: plot3d(lambda x,y : fermi(x,y,0.2,1), (x,-2,2), (y,-2,2)) > > Does this look fine? Yes, the "pure python" way works. But defining the function that way, I loose the ability to perform symbolic manipulations with it, don't I? -- 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] Re: plot3d and expression evaluation
On Mar 12, 5:39 pm, stefan wrote: > I have encountered a somewhat strange problem ... Ah, I got it, the max function is evaluated during the definition of the function, look: sage: fermi(x,y,d,L) = 1 - 1/( exp( ( max(abs(x),abs(y))-L) /d) + 1) sage: fermi (x, y, d, L) |--> -1/(e^(-(L - abs(x))/d) + 1) + 1 especially: sage: max(abs(x),abs(y)) abs(x) There is no y any more! H -- 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] Re: plot3d and expression evaluation
On Mar 12, 5:39 pm, stefan wrote: > Basically, I am trying to plot this expression: > > #sage> fermi(x,y,d,L) = 1 - 1/( exp( ( max(abs(x),abs(y))-L) /d) + 1) I'm not sure but there might be a bug or problem evaluating the expression. Anyways, going the "pure" python way works: sage: def fermi(x,y,d,L): return 1 - 1/( exp( ( max(abs(x),abs(y))- L) /d) + 1) sage: plot3d(lambda x,y : fermi(x,y,0.2,1), (x,-2,2), (y,-2,2)) Does this look fine? H -- 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
