« first day (2098 days earlier)      last day (2383 days later) » 

2:07 AM
What would you say the best practices with functions involving units are? I'm trying to get a value for a transfer function (complex and unitless) but input values have units (Hertz, Farads, Henries, etc), and when I put stuff into mathematica it calculates units correctly but refuses to convert anything and give me the dimensionless result
I tried UnitConvert and UnitSimplify, but they don't seem able to convert the units and realize the quantity is unitless
 
 
2 hours later…
3:55 AM
@inkyvoyd I think sometimes you need to be a bit explicit about transformations. Some transformations like "Wavenumbers" to "Megahertz" have implicit conversion factors that Mathematica wisely ignores. In general if you have big data I'd say you want to take all your data with units to start, convert everything to a standard unit system, and do your computation solely with numerics.
For the general community's perusal, here's a somewhat disappointing but wholly unsurprising thing. I built a service connection to the GitHub jobs API (because why not?) and decided to try to see what sort of Mathematica jobs exist on it.
There are exactly 0
If you want to try this for yourself, run the following:
(* Install the connection *)
PacletManager`PacletInstall["ServiceConnection_GitHubJobs",
 "Site" ->
  "http://www.wolframcloud.com/objects/b3m2a1.paclets/PacletServer"
 ]
$so = ServiceConnect["GitHubJobs"];

$so["JobsListing",
 "description" -> "Mathematica"
 ]

{}
$so["JobsListing",
  "description" -> "Java"
  ] // Length

48
 
thanks for the suggestion. I don't really have a whole lot of data to work with, so the unit conversion feature is more of a convenience (and error checker) than anything else, but seeing as I'm more or less an amateur to mathematica, it's proving to be a major headache
 
@inkyvoyd the implementation is admittedly a bit on the clunky side
 
@inkyvoyd what explicitly are you trying to do? Generally UnitConvert should do what you need.
 
Basically just trying to do some calculations for solid state physics (photodiodes) wherein there are a lot of constants, and a few variables; everything has units
 
Could you give a sample formula?
 
4:06 AM
Well, I sort of answered my own earlier question in the sense that I had a typo; right now I'm trying to wrangle NSolve through an equation. The formulas I'm working with are pretty long and seeing as I'm an amateur I'm not sure I've written them in an understandable way; at this point I am wondering if learning how to use the built in units feature is really worth my time
 
for NSolve[] it's usually much less of a headache to make everything unitless
 
yeah; I'm more or less convinced at this point that if I wanted to spend time learning about mathematica picking up a book and just reading through it would probably be more helpful than trying to search around for solutions to small things... it's a pity this platform is so esoteric for more advanced operations
 
Well, this is a system that rewards mastery.
The "esoteric" stuff usually have good reasons why they were made so.
(operative word "usually")
 
4:40 AM
1
Q: Mathematica.SE User Inteface Problems

bbgodfreyYesterday, I noticed that my Mathematica.SE user page sometimes was different, depending on how I reached the page. Today, after I rebooted my computer, the appearance of nearly all Mathematica.SE pages changed. For instance, the top part of my user page now looks like this and the Ask page ...

 
 
6 hours later…
 
1 hour later…
11:27 AM
@b3m2a1 Inaccessible. Probably you need to redeploy with Permissions->"Public"
 
FYI - a trick:
Simulating GeoNearest with GeoEntities & GeoDisk
For speed
And no timeouts
 
12:03 PM
0
Q: Issue with code readability due to font style

SzabolcsThere is a problem with code readability due to the font used on this site. It is not clear if the font is the same on all platforms, but on my computer (OS X) f[a_, α_] := α + a^2 renders as Here it is in double size. The problem is that the italic and the upright versions of a look co...

 
 
3 hours later…
2:42 PM
many questions are posted using Subscripts notation, and I can't read them and when I copy/paste the code to the notebook, they remain in the explicit Subscript notation so still hard to read. Is there a way around this? How do others read these questions? How to convert it to 2D math so at least one can read it? I can ask this on meta if needed. I can't read many posts code since everyone seems to use subscripts.
 
@Nasser Try Ctrl-Shift-N (not sure)
 
@Szabolcs Thanks! That worked. I never knew about this one, will add it to my cheat sheet now :)
Is there a way to also make the posted code show the same way as in 2D in the notebook? So it is easier to read in the post itself without having to copy it to the notebook and then do ctrl-shift-N each time to see it?
 
3:37 PM
@RolfMertig it should be public. I think the site file just got corrupted somehow. I fixed it I think, but I'm not entirely sure how.
 
 
2 hours later…
5:10 PM
StackOverflow now has tags where questions seem to get more downvotes than upvotes, in total. I keep scrolling through and it's either 0 votes or negative. Positive votes are few, and almost nonexistent during certain times of the day.
 
5:39 PM
@Szabolcs I love it :(
 
 
2 hours later…
7:50 PM
@Nasser Did you contact WRI about this?
 
@SjoerdC.deVries sorry, about what?
 
@Nasser Just press the left arrow button to the left of my message ;-)
 
@SjoerdC.deVries oh, I see now the left arrow thing. No I have not. I have not yet finished the report. I know why it happens, it is becuase few integrals now generate hug result compared to before. But need to find those integrals which are the problems. There are few integrals which generate huge results in 10+ versions compared to 9. I need to first find those. I still have not done this. These few integrals skew the mean of the overall test.
I posted 2 such ones above.... on this chat page as an example. But there are many more...
 
 
2 hours later…
9:57 PM
Hi. Can anyone help me get a package to install and work on Mathematica 11, on a Mac? I'm not sure what's the problem.
 
What's the package?
 
What's the problem?
 
Any package, really. I have the .m file of the package with me. Now do I have to place the file in a specific folder?
I've tried following the steps I saw online and I always get a "$Failed" message
 
You can run Get on the file
Or put the directory on the $Path and do <<PackageName`
 
Okay so, here's what I did. In a notebook, I typed "$HomeDirectory" which then gives me a specific location. I go to that location and paste the .m file of the package. I go back to the notebook and evaluate "<<packagename`"
And it works
This is good. But I don't want to put all the m files in that home directory. I want Mathematica to automatically look for m files in a specific folder where I keep the m files. So I'd want Mathematica to search for all the folders and subfolders in the specific folder I choose.
How do I do that again?
 
10:15 PM
@Avantgarde Folders and subfolders is a bad idea since packages itself contain folder structures.
 
@halirutan That's right. So you suggest pasting all the individual m files in one single folder?
 
Usually, a package contains not only one .m file. It has a certain structure.
You can inspect how Mathematica does it.
 
Hm alright
 
When you look at the $Path variable, you see for instance
/usr/local/Wolfram/Mathematica/11.2/AddOns/Packages
Take a look at this folder.
@Avantgarde The correct place where your own packages should go is
FileNameJoin[{$UserBaseDirectory, "Applications"}]
There you can put your .m files or package-directories and they will be found by Mathematica.
 
Yes!
Thank you very much. It all falls into place now. :)
 
10:29 PM
@Avantgarde No problem.
@Avantgarde Just to tip you off: Mathematica searches all folders that are in $Path. You either can add something to it and you make sure you do this on startup with your init.m so that it is always available, or you just put your packages into a place that is searched anyway.
 
11:00 PM
@halirutan Thank you for the tip
 

« first day (2098 days earlier)      last day (2383 days later) »