On Nov 14, 2008, at 3:08 PM, Jason Grout wrote: > William Stein wrote: >> I did try pasting that example into sagenb.org and it gives >> some weird errors involving _fast_float. Jason Grout -- maybe >> you could look at why your interact appears broken? > > Robert Bradshaw: I've asked a question at the bottom of this email to > you about partial function evaluation of fast_float functions... > > Okay, I've updated the code to be smarter. The code ended up calling > maxima a *lot* for what basically was partial function evaluation. > Instead, I switched it to use the functools.partial class to partially > evaluate a fast_float function. Apparently I triggered the surge > protection on the wiki and so cannot post the update there. It is > here: > http://sagenb.org:8000/home/pub/69 and also just in case sometime > in the > future, the public notebook server goes down, here is the code so it's > archived on the list: > > var('u v') > from sage.ext.fast_eval import fast_float > from functools import partial > @interact > def trans(x=input_box(u^2-v^2, label="x=",type=SR), \ > y=input_box(u*v+cos(u*v), label="y=",type=SR), \ > t_val=slider(0,10,0.2,6, label="Length of curves"), \ > u_percent=slider(0,1,0.05,label="<font color='red'>u</font>", > default=.7), > v_percent=slider(0,1,0.05,label="<font color='blue'>v</ > font>", > default=.7), > u_range=input_box(range(-5,5,1), label="u lines"), > v_range=input_box(range(-5,5,1), label="v lines")): > thickness=4 > u_val = min(u_range)+(max(u_range)-min(u_range))*u_percent > v_val = min(v_range)+(max(v_range)-min(v_range))*v_percent > t_min = -t_val > t_max = t_val > g1=sum([parametric_plot((i,v), t_min,t_max, rgbcolor=(1,0,0)) > for i > in u_range]) > g2=sum([parametric_plot((u,i), t_min,t_max, rgbcolor=(0,0,1)) > for i > in v_range]) > vline_straight=parametric_plot((u,v_val), t_min,t_max, > rgbcolor=(0,0,1), linestyle='-',thickness=thickness) > uline_straight=parametric_plot((u_val, v), > t_min,t_max,rgbcolor=(1,0,0), linestyle='-',thickness=thickness) > > (g1+g2+vline_straight+uline_straight).save > ("uv_coord.png",aspect_ratio=1, > figsize=[5,5], axes_labels=['$u$','$v$']) > xuv = fast_float(x,'u','v') > yuv = fast_float(y,'u','v') > xvu = fast_float(x,'v','u') > yvu = fast_float(y,'v','u') > g3=sum([parametric_plot((partial(xuv,i),partial(yuv,i)), > t_min,t_max, rgbcolor=(1,0,0)) for i in u_range]) > g4=sum([parametric_plot((partial(xvu,i),partial(yvu,i)), > t_min,t_max, rgbcolor=(0,0,1)) for i in v_range]) > vline=parametric_plot((partial(xvu,v_val),partial(yvu,v_val)), > t_min,t_max, rgbcolor=(0,0,1), linestyle='-',thickness=thickness) > uline=parametric_plot((partial(xuv,u_val),partial(yuv,u_val)), > t_min,t_max,rgbcolor=(1,0,0), linestyle='-',thickness=thickness) > (g3+g4+vline+uline).save("xy_coord.png", aspect_ratio=1, > figsize=[5,5], axes_labels=['$x$','$y$']) > print jsmath("x=%s, \: y=%s"%(latex(x), latex(y))) > print "<html><table><tr><td><img > src='cell://uv_coord.png'/></td><td><img > src='cell://xy_coord.png'/></td></tr></table></html>" > > > > Robert, can we make partial function evaluation part of fast_float? > That way, given the following: > > var("u,v") > x=u^2+v^2 > xuv = fast_float(x,'u','v') > > the following are equivalent: > > xuv(2)(3) > > and > > xuv(2,3) > > > Of course, right now, we can do this (with a slight performance > penalty) > by doing: > > import functools.partial > > functools.partial(xuv,2)(3) > > > My whole reason for doing this (to avoid expensive maxima calls) is > disappearing soon, so maybe it's not worth the effort, especially > since > functools.partial provides a standard python way to get this.
It certainly could be done, but I don't know how worth it it would be. What notation should we use. (I'd much rather have an error when one enters an incomplete list of arguments). - Robert --~--~---------~--~----~------------~-------~--~----~ To post to this group, send email to sage-support@googlegroups.com To unsubscribe from this group, send email to [EMAIL PROTECTED] For more options, visit this group at http://groups.google.com/group/sage-support URLs: http://www.sagemath.org -~----------~----~----~----~------~----~------~--~---
