« first day (2800 days earlier)      last day (31 days later) » 

4:07 AM
@Szabolcs ListPlot[data]/.Point[p]:>Point[p, VertexColors->getColor/@p]?
That's what I tend to do and it's generally fast
 
 
4 hours later…
8:04 AM
Is it possible to change the order of an already created InterpolatedFunction?
 
8:25 AM
@Kuba The interpolation order? You would have to re-create it from the source data (which is embedded in the function)
 
@Szabolcs yes, I can do that, I just thought there might be something easier.
btw, is there InterpolatingFunction[...]["Domain"] documented? I just tried it and it worked but I didn't know before.
 
My guess is that there isn't, as all the coefficients need to be recomputed, i.e. changing the order is the same as re-creating it.
Maybe user21 knows.
@Kuba None of that is documented. Try if["Methods"] (not "Properties"!!)
@Kuba Also try if["MethodInformation"["Domain"]]
It's a little unusual because unlike almost all other similar objects, it uses Methods instead of Properties.
 
Interesting, thanks.
 
I think it might be because it's actually older than other similar things.
 
Anybody else having constant problems with wolfram cloud? 50% when I open Wolfram Desktop it won't sign me in correctly and ask me to change my $CloudBase (to what it is currently set to). Lots of control-= evaluations fail first time as well.
 
 
9 hours later…
6:01 PM
Lots to read here, nice! Been studying and teaching like some kind of graduate assistant or something, howdy hey everyone!
Quick Q: when one makes a Table or a ParallelTable to be specific, what is the order of evaluations that occur? I'm wondering if it is each of the outermost rows that are computed first, or just some combination of what is done first, then the next part, and so-on. Does the ordering occur at the end or beginning? Most precisely I am wondering if ParallelArray might be faster, as it wouldn't go row-by-row as I think Table does, but index-by-index?
Does maintaining an unused list take up time in an evaluation? Or would some combo of Scan and Table result in a faster evaluation?
 
 
2 hours later…
8:34 PM
@CarlLange I think it's still a year or two out before the cloud is a real semi-robust product
 
9:26 PM
Finally some questions about chemistry on here, many months after 12.0 release......
 
 
2 hours later…
11:55 PM
Hello,

Does anyone know of a good way to specify boundary conditions at infinity for solving partial differential equations (specifically reaction-diffusion equations)? I have been trying to just specify the condition at a very large value for the spatial variable, but I get warnings about the step sizes, and I think it makes adds time to numerically determine the solution.
I have tried searching previous posts, but it seems like the answers always become "Well we know the solution is this, so just solve the diffEQ this way instead." But I am looking at some complicated, coupled, non-linear equations that probably don't have a closed form solution. I need to numerically solve the equations
 

« first day (2800 days earlier)      last day (31 days later) »