everyone fall asleep in here?

Nope, of course not.

Question: Assume a very simple Dataset

ds = Dataset[Table[Association["value" -> RandomReal[]], 10]]

What I want is to separate this Dataset into values that are smaller and larger 0.5 and I need a list {{small1, small2, ...}, {large1, large2, ...}}. I can do this with

ds[GroupBy[#value >= .5 &] /* KeySort, All, #value &][Values] // Normal

ds[GroupBy[#value >= .5 & -> (#value &)] /* KeySort /*  Values] // Normal

@halirutan like that ^ ?

@Kuba Simple examples sometimes have the flaw that they are too simple :) The problem is when I have something like this

ds[GroupBy[#value >= .5 &] /* KeySort, All, {#value, #value^2} &][Values] // Normal

Basically, my Dataset is much more complicated and I need to do serious operations on them.

@halirutan then GroupBy[#value >= .5 & -> ({#value, #value2} &)]?

I'm not sure I got your update

@Kuba That seems to work but I'm not sure I like it. Especially with complicated group conditions and a larger calculation.. hmm.

Nevertheless, thank you very much

@halirutan np. so, which step is supposed to grow up to make it ugly. (#value&)?

( ′◔ ‸◔`)

the tutuorial of package devepoment of Wolfram Mathematica is REALLY unclear，(┬＿┬)

@xiang :D:D:D sure it is, but not only to you so that many question are already asked and answered here.

oh, you just comment on my quetion "How do I activate the infobutton that links to the documentation page for package functions"

@xiang just query packages // SortBy["votes"] and you will find many useful things.

mathematica.stackexchange.com/questions/tagged/…

Yes, I did ;)

I have a quetion that is it called "information button" or "info-button" or something else?

Does it matter?

maybe the wrong name will make us search not as easy as the right name

Maybe but when it becomes documented the documented name may not be the same as the one established by community so I'd not care.

Already asked but unfortunately no answers.

(┬＿┬)

A lot of useful info here: mathematica.stackexchange.com/a/132781/12

@xiang mathematica.stackexchange.com/a/132960/12

@Szabolcs " Every public function will have a usage message in this section." - I'd rephrase it to it "every public function has to be mentioned here and it is a common practice to put here usage messages" or something.

@Szabolcs and link to mathematica.stackexchange.com/q/125579/5478

@Kuba I did that on purpose to keep things simple and easy. I'm writing a note now to explain that what I wrote in the guide isn't strictly true.

@Szabolcs I know but I'd go with "vague and linked" rather than "not true for compactness reasons"

But this is only a suggestion

@Kuba I know what you mean, but this time I decided go with not true on purpose, will be clarified later. It's really difficult to do this well. I'll try to rephrase a bit to make things "more true".

@xiang Perhaps you can explain why you think that the tutorial is unclear.

ok

as we konw，when we move our mouse curse into a Symbol like Map

It

It will pop up a window with a circle icon "i"

That's not a beginner topic.

I suggest you leave that for later.

if we click the "i", we will move to the ref/Map

in the document center

In fact I think it is not possible to provide this "i" symbol for your own functions. But it is possible to integrate with the documentation center, you just won't see the "i".

Why don't you believe me when I say it is internal feature not ready for users?

I have already integrate my application with the documentation center

if I search my self defined function in documentation center, it will show my tutorial ,guide and ref

@Kuba Maybe you are right. The Lie Algebra Package - LieART also has not the "i" button for it's own function

Only builtins have that button.

嗯嗯

thanks

@xiang Well, it is not a good idea to trust strangers but you can always ask WRI Support and get info that will fit as an official answer.

嗯~ o(￣▽￣)o you are right

Shouldn't we have a close reason which just says "not about Mathematica", without further explanations? mathematica.stackexchange.com/q/132957/12

There are too many obviously off topic questions and many are of low enough quality that they wouldn't be received well on math.se either.

@Szabolcs Off topic / Out of scope is for this I suppose. I think it should be interpreted "it is out of scope or it requires advice from WRI..." but without "or" this really sounds like "and" ;)

Why I get so many Close?

Can anyone give a reanson?

Or advice?

Is there a good way to find posts that were closed as duplicates of other posts? Or a specific post. In particular, I'm looking to collect examples of people confused about Abs'[x] in Mathematica, similar to mathematica.stackexchange.com/questions/120577/…

@Searke How about searching for is:Q duplicate:yes Abs?

Or even

is:Q duplicate:yes title:"Abs"

@halirutan thanks. I guess I never looked at the search options.

@Searke no problem

@yode The original post is more of a project than a specific question. It's a cool problem.

@Searke All the duplicates appear in the list of linked posts. But not all linked posts are duplicates.

@Szabolcs I'm dealing with a group of people this morning who don't believe me that confusion about Abs'[x] is a common enough problem to warrant documentation

@yode I think it's an interesting problem. Did you look at my link?

google.fr/…

google.fr/…

@Searke I found a very high number of posts with the above two google searches. Most do seem relevant. I hope it will help.

Thanks.

Agreed to tag this as bug? mathematica.stackexchange.com/q/132689/12

@Szabolcs Of course,I have.Thanks for your link,but I'm a Matlab newbie :)

But they should have a description of the method there

@yode @Szabolcs I've voted to close the "Triangle question" not because it isn't an interesting/useful/challenging question. But the question is not written as such. The reason is that the question is written for a single, arbitrary, artificial, unrealistic (what real life process would have points generated by a uniform distribution in a triangle and a uniform distribution in a polygon that excludes the triangle?) set of data points.

My requests for what the problem to be solved is in general or what processes generate such data points have been ignored or at least thought irrelevant. The second issue is that some of assessment of the process on multiple datasets is required to have any idea if the method works in general and how well it works.

@yode How do you know they ignore them? Maybe it is not easy to solve the problem. In a few cases, they disagree with users about the problem. Sometimes there is internal disagreement about what fits with the overall design, perhaps. Sometimes a bug has a low priority and they have finite resources. I think it probably makes sense, but I also respect users who disagree with the priorities the company sets.

Slapping 3 lines around the triangular cluster of points by eye and then figuring out the intersections to define the triangle would be just as "objective" and possibly better and quicker. But clearly, there must be multiple datasets in mind. I'm simply trying to get @yode to give more details.

@JimBaldwin As your "arbitrary" request,I just use short word to describe problem to avoid make grammar error in my sentences.And you second request,It is sure a concept about how to calculate ternary phse diagram emerge my mind,so I use uniform distribution to simulate those data.Maybe it will give you a unreasonable thinking.Lastly,it's a awkward reason that-I cannot understand your English fully.

@MichaelE2 Yes,I'm sure that.Such as the RelationGraph's unexpected behavior,I have post in MSE long long ago.I realize that not useful,so I send a Email to WRI,as the response of response Email.I think the official don't know those behavior fully.

@halirutan Another option would be to convert the final Values into an ascending operator:

ds[GroupBy[#value >= .5 &] /* KeySort /* Query[Values],
  All, {#value, #value^2} &] // Normal

or

ds[GroupBy[#value >= .5 &] /* KeySort /* (Values[#] &),
  All, {#value, #value^2} &] // Normal

Anyone know if there's a variant of Skip that plays nicely with the datatypes for BinaryRead? I have a code that seems to work but it triggers:

Read::readf: UnsignedInteger16 is not a valid format specification. >>

@dionys Maybe SetStreamPosition?

Do you think this is too close to being off topic? Should it go on Math.SE? It's about how a feature that several CAS have may be implemented. It's for my computer algebra education :) mathematica.stackexchange.com/q/133019/12

@Szabolcs I just wanted something that plays nicely with the binary functions, so I wouldn't have to add a hacky hard-coded translation table. I'm surprised Skip isn't already built to handle binary type specs ... or alternatively, that there isn't a builtin BinarySkip to complement BinaryRead.

I agree.