Wolfram Mathematica

rm -rf

12:04 AM

@VitaliyKaurov Heh, yet another link... :)

Vitaliy Kaurov

@rm-rf ;=) you mean to closure or to community ?

probably all of them..

rm -rf

@VitaliyKaurov Oh, I meant to Clojure... I like the links to Community :) Lets me (and others that aren't active there) know what's happening there :)

Vitaliy Kaurov

i was surprised by that Clojure project - you know about other ones ?

@rm-rf

rm -rf

@VitaliyKaurov There's MATLink for MATLAB. I believe Seth Chandler had a Scala link as well. There's an undocumented python link in the extra packages in $InstallationDirectory, but I haven't used that.

I remember seeing a video with Seth demoing Scala link, but I'm not sure if there's a public package

Vitaliy Kaurov

interesting.. what does it add to JLink ?

@rm-rf

sorry for asking - i am not sure if you know the details

rm -rf

12:18 AM

@VitaliyKaurov Sorry, I might have gotten it wrong there. You're right; it seems like Seth used JLink to connect to Scala and Clojure... wolfram.com/broadcast/… (if the abstract is correct)

Vitaliy Kaurov

@rm-rf yes yes that's it. But that Clojure project king of jumped from nowhere - i event not sure who is developer - i bet he is here on MSE

even*

rm -rf

@VitaliyKaurov Oh, he is here (and an early user of the site as well), although I can't reveal who it is since they've used an anonymous username :)

Vitaliy Kaurov

@rm-rf no problem ;-) was that project discussed on MSE in chat before?

rm -rf

No, not the clojure thing. That's totally new and unannounced

Vitaliy Kaurov

Btw, in terms ow links to W.Community - I forget about real cool things, just post always moving stuff ;0 Here is one I really like:
http://community.wolfram.com/groups/-/m/t/178325

@rm-rf

BUt it is a long reading, sorry for abuse ;)

nvm if complex systems is not yr cup of tea

was my study subject

rm -rf

12:37 AM

@VitaliyKaurov I've long forgotten most of that stuff (only had it in one course in undergrad), but that does look interesting! As I said before, I always enjoy the links to WC, especially the cool stuff and interesting discussions :) I think it's also important to let people know that there's a place outside of mma.se where they can have a "discussion" with other mma users (something that's expressly disallowed here).

Vitaliy Kaurov

@rm-rf i entirely agree. I love how MSE dissects problems quick with surgeon precision. but i also see so many projects scattered around all over the web or in chats that just disappear or remain unknown. it's nice to have a nice chill out place where folks can just share. gather em up, classify

bb in a bit

rm -rf

@VitaliyKaurov We tried to do that here: meta.mathematica.stackexchange.com/q/428/5, but the hard truth is that Mathematica users, as a community, don't really develop and share/use packages... everyone makes their own little package for their needs (or clumsy solutions using multiple notebooks) and that's it. There's a lot of "reinventing the wheel" going on in the community.

In a lot of ways, I think it's a result of how WRI views Mathematica and its position in the programming/computational science universe and how it treats its developer community... I don't know if this situation will change anytime soon.

Vitaliy Kaurov

1:47 AM

@rm-rf I think cloud will help to change the things. btw to complete the Clojure links:
http://clojuratica.weebly.com/
http://xahlee.blogspot.com/2013/11/wolfram-language-and-lisp.html

novice

2:59 AM

How to reshape a list according to another list, like this: A={1,2,3,4};B={{q,p},{r},{t}}; Reshaping A according to B is A={{1,2},{3},{4}}

And reshape according to the numbers, such as A={a,b,c,d,e,f};B={2,1,3};Reshaping A according to B is A={{a,b},{c},{d,e,f}};

Oleksandr R.

31

A: Partitioning with varying partition size

The core solution If I understand your question I previously wrote a function for this purpose. The core of that function is: dynP[l_, p_] := MapThread[l[[# ;; #2]] &, {{0} ~Join~ Most@# + 1, #} & @ Accumulate @ p] Version 8 users have Internal`PartitionRagged which has the same syntax for ...

novice

3:17 AM

Oleksandr R. Thanks

novice

6:52 AM

I have some experience with Matlab. The command "clear all" erase all variables and their values. I am wondering what is the difference between "quit the kernel" and "Clear" in Mathematica? Is there a shortcut key or function like "Clear" to quit the kernel?

Rojo

9:45 AM

@novice Exit, Quit?

Calle

10:39 AM

@novice The example in the documentation for ClearAll, ClearAll[Evaluate[Context[] <> "*"]], comes pretty close MATLAB's clear all. It will erase all variables and functions (all symbols, that is) in the current name space/context (Global usually, if you don't use contexts then just imagine it will erase all symbols that you've defined).

rm -rf

1:59 PM

@Calle ClearAll@"`*" is easier to remember (and shorter) :)

Sektor

3:33 PM

@rm-rf Got a second ?

rm -rf

4:17 PM

@Sektor I'm here now...

Sektor

4:37 PM

@rm-rf Got a moment to answer another question :D

rm -rf

@Sektor What's up?

Sektor

@rm-rf Hi ! I have a question if you have the time :)

ContourPlot[
Re[-(1/2) I E^(-(1/2) (1 + \[Omega])^2 -
1/2 I t (2 + \[Omega])) (-1 + E^(
2 I t + 2 \[Omega]))], {\[Omega], -\[Pi], \[Pi]}, {t, 0,
20 \[Pi]}, ImageSize -> 500]

This produces perfectly fine plot, but was it possible to exchange the axes and preserve the orientation of the plot

I can easily change the places of the two plot dimensions, but that will rotate the plot

rm -rf

@Sektor You mean, just change the tick labels?

Sektor

nah, not just the ticks, that's doable :) I want to preserve the plot dimensions and make the plot facing me i.e. as it is in the example

rm -rf

@Sektor I didn't understand what you meant by "exchange the axes"... do you want the x axis to read from 0 to 60?

Sektor

4:45 PM

@rm-rf No, again that's doable :) I want the time axis to be horizontal

and the freq to be the vertical while preserving the "normal" orientation of the plot

rm -rf

Hmm... is that even possible :D

Sektor

@rm-rf That's what I am asking :D

rm -rf

As I understand it, you don't want this, right? i.stack.imgur.com/xME6N.png

Sektor

absolutely correct, I don't want it :)

I am trying to increase the readability and what better way than just rotate the plot :D

rm -rf

Well, then I don't think it is possible, because that's just how your function is :)

Sektor

4:47 PM

buuut this way I am just changing the axes

Yup, prolly, but then there's an error in the paper I am reading right now :D: D :D

rm -rf

@Sektor Do you have a link to the paper/screenshot of the figure?

Sektor

I can post a screenshot

rm -rf

That'll do

Sektor

Nope, no need to, I found my error, If you are still interested

it is a comparison between gabor and wdf :)

I can post it

rm -rf

@Sektor If it's resolved, then no need :)

Sektor

4:52 PM

@rm-rf Superb, thank you for your time :)

rm -rf

No problem :)

Sektor

@rm-rf Btw, what are the color schemes you use the most ?

rm -rf

@Sektor I very rarely use the built in colorschemes for a publication plot... I create my own depending on what is being plotted.

Sektor

@rm-rf Me too, I was interested in the one you used :D

rm -rf

For quick work, I like sunset colors (usually reversed), deepsea colors, sometimes beach colors... or, plain ol' rainbow/lake colors :)

Sektor

4:56 PM

@rm-rf Specifically for this contour plot

@rm-rf Thanks again :)

Jacob Akkerboom

5:45 PM

Who will give me bounty if I write the fastest answer to the question "Partitioning with varying partition size", using a funky new technique and LibraryLink? :P Nobody likes LibraryLink :(. (first I will clean up my mess, don't look for an answer from me)

31

A: Partitioning with varying partition size

The core solution If I understand your question I previously wrote a function for this purpose. The core of that function is: dynP[l_, p_] := MapThread[l[[# ;; #2]] &, {{0} ~Join~ Most@# + 1, #} & @ Accumulate @ p] Version 8 users have Internal`PartitionRagged which has the same syntax for ...

WReach

6:31 PM

@halirutan Here's something that might get you started...

Needs["CompiledFunctionTools`"]

"Op Codes"
Cases[
  Names["CompiledFunctionTools`Opcodes`CC*"]
, n_ :> {ToExpression[n, InputForm], StringReplace[n, ___~~"`CC":>""]}
] // Sort // Grid

"Function Call Names"
DownValues[CompiledFunctionTools`Opcodes`Arg1Name] /. (_[_[n_]] :> s_) :> {n, s} // Sort // Grid

"Instruction Interpretation"
Block[{$Context = "CompiledFunctionTools`Private`"}
, DownValues[CompiledFunctionTools`Private`getInstruction] /. HoldPattern -> HoldForm //
  Column //
  Print
]

halirutan

6:44 PM

@WReach I found that already, but thank you anyway.

Jacob Akkerboom

7:04 PM

Shit shit, I wanted to make a self-answered question, but I then I decided I didn't want to answer it after all. When I finished the question, I clicked "discard draft" at the bottom of the page to discard the answer, but the entire Q&A was gone. Is there any way to retrieve this or is it gone forever?

frustration

rm -rf

7:22 PM

@JacobAkkerboom Frustration? THIS is frustration:

3

Jacob Akkerboom

@rm-rf hehe, call me overprepared, but I have that exact .dmg on my harddrive ;)

or at least 9.0.0

sec

Ah shit, I only have Mathematica_9.0.0_OSX.dmg :P

Hehe subtle ;)

halirutan

7:51 PM

@rm-rf Why? It gives you a lot time to think about interesting stuff. Tip: Start such installation routines on Monday at work!

@JacobAkkerboom You want to use LibraryLink? Please remember the exact specification of the question:

> The length of the k'th partition is a function f(k).

You found a way to provide function pointers?

Jacob Akkerboom

@halirutan I figured that I'd probably go for a library that can do poth

hmm function pointers? why? Oh no I was going to calculate f in Mathematica-proper :P

Of course if you can compile f yeah maybe you should compile f, but I guess it could be anything :)

So the input of the librarylink function would be a packedarray of integers which are the length of the "elements" of the partition

as well as the list of values

and the output is a slight modification of the list of values, with seperators

I imagine that is very fast. I was going to do something similar in some project, but I forgot where

Ah yeah I remember

I was going to write a function to do FactorInteger/@Range[n] faster. The result is a ragged array, I figured this approach using SplitBy would be fastest

cu guys

Vitaliy Kaurov

9:56 PM

1

A: Make a scalable Christmas Tree

Mathematica ASCII I really like ASCII art, so I add another very different answer - especially so it's so short in Mathematica: Column[Table[Row[RandomChoice[{"+", ".", "*", "~", "^", "o"}, k]], {k, 1, 35, 2}], Alignment -> Center]

bobthechemist

@VitaliyKaurov Nice

Wolfram Blog

10:41 PM

Wolfram Blog: Launching the Wolfram Connected Devices Project

posted on January 06, 2014 by Stephen Wolfram

Connected devices are central to our long-term strategy of injecting sophisticated computation and knowledge into everything. With the Wolfram Language we now have a way to describe and compute about things in the world. Connected devices are what we need to measure and interface with those things. In the end, we want every type of [...]

Transcript for