« first day (2654 days earlier)      last day (1840 days later) » 

Jason B.
Jason B.
1:16 AM
@Szabolcs Here you can do Molecule[bucky] /. "Single" -> "Aromatic" - I should have put GraphMolecule in the chemistry context but it's buried. In general you can do Molecule[ _Graph] but it will make any degree-1 vertex hydrogen and anything else carbon, and all single bonds.
 
Jason B.
Jason B.
1:36 AM
@b3m2a1 I need to check on this
you can see the difference between these: `Graph3D@CanonicalGraph@GraphData[{"Fullerene", {60, 1}}]`

`Graph3D@CanonicalGraph@ChemicalData["FullereneC60", "StructureGraph"]`
here it is - look at the distribution of 5-membered rings
If I take the graphdata example, `mol = Molecule[GraphData[{"Fullerene", {60, 1}}]] /.
"Single" -> "Aromatic";`
then here are all the 5-membered rings highlighted:
but in a normal buckyball they aren't grouped like that:
`mol2 = Molecule[
Entity["Chemical", "FullereneC60"][
EntityProperty["Chemical", "SMILES"]]];`
 
 
3 hours later…
Meta
Meta
4:45 AM
0
Q: Post contains a code that is not properly formatted

Mee Seong ImI am trying to post this question on Mathematica StackExchange but I am getting an error that my code is not properly formatted. I double-checked every line for some time now, but I cannot find an error. Can someone help me to see what is wrong with my post? This is not the first time that this ...

bug code-formatting
 
 
2 hours later…
1010011010
1010011010
7:00 AM
Parallelize seems to be doing something strange to my code
I am aware that I should change variable names with subscripts to names without them
So I have done so,
Now my parallelize code gives strange errors, for example "Nonatomic expression expected at position 1 in AppendTo[\
\[ScriptCapitalE]\[LetterSpace]list,-4.839705861954373`]"
I can enclose the value inside braces but it doesn't matter much
The heads of the expressions inside: are a mix of List and Symbol
I think it's a good idea to first make sure that these are both Attributed "List"
How do I do this?
The "unevaluated" expression that gives that -4.83 is: `-(q[p])^2 +
Sum[2.*d\[LetterSpace]list[[i]]*\[Lambda]\[LetterSpace]list[[i]]*
L + (d\[LetterSpace]list[[i]])^2, {i, j\[LetterSpace]list}]/(L^2)`
Lambda_list is created using Findroot, like so:
\[Lambda]\[LetterSpace]list\[LetterSpace]symb /.
FindRoot[BE\[LetterSpace]array, \[Lambda]\[LetterSpace]list\
\[LetterSpace]symb\[LetterSpace]appended]
where Lambda_list_appended is a function created through "Map"
Now if I do the exact same thing in d_list, the d_list is now a Symbol instead of a List???
I'm utterly baffled
 
Carl Lange
Carl Lange
@1010011010 Might be good to add a question on the main site.
2
 
b3m2a1
b3m2a1
7:34 AM
Example of how I could foresee myself distributing multipage Mathematica documents
Doesn't have any explicit reference to a stylesheet although I probably should make one for it that I just steal all the styles from to cook them into the notebook.
 
b3m2a1
b3m2a1
7:47 AM
@Szabolcs Seen this?
7
Q: Can I implement something like the old documentation help browser?

b3m2a1Old versions of Mathematica had a help browser that was more click through and structured. How can I reimplement something like that for the new (since v6) docs?

customization documentation
 
 
2 hours later…
Carl Lange
Carl Lange
9:19 AM
@b3m2a1 This looks really nice.
 
 
5 hours later…
user929304
user929304
2:42 PM
@MichaelE2 Hi, I had a quick question about compact notation to normalize vectors if you're around
we have an array u of vectors (so each element of u is a vector, e.g. {10.1,2.2,30.}). How can we apply Normalize to u such that it gets applied element wise to all its vectors?
@MichaelE2 sry, figured it out ;) using Normalize /@ u
 
 
1 hour later…
Kuba
Kuba
3:50 PM
@Searke I overwrote your edit because that bug is not fixed in 12.0.0. The exact example is enough to reproduce it. mathematica.stackexchange.com/q/95351/5478
 
 
2 hours later…
halirutan
halirutan
5:29 PM
@Szabolcs @b3m2a1 @ArnoudBuzing @Kuba I talked with Arnoud in private about the change of our SE header image to include "Wolfram Language" and the public bug-tracker, but my opinion matches that of Szabolcs. The header image is a community decision and I would do a meta post and if the community agrees, I could talk to our designer.
 
b3m2a1
b3m2a1
@halirutan sounds good to me
 
halirutan
halirutan
@Szabolcs What do you think if we bundle our pros and cons for the public bugtracker to give Stephen a list of the core-thoughts of independent developers and long-time users about that matter?
2
Of course, everyone is welcomed to give their two cents. I believe the biggest fear which needs to be destroyed is that a public bucktracker is bad and will harm the image of Wolfram Inc.
 
yode
yode
5:47 PM
1
Q: How to make an API with CloudDeploy to download an image into the local disk?

yodeActually related question 127248. But in that post, the answer from Kuba suggest provide a URL for API-user. But I still hope the API can output the file into the local disk directly. Is it possible? As the Kuba's comment, the browser cannot write the local disk otherwise it will lead to malici...

cloud-deploy api
Can we use the API by wolfram deploying to download a image?
 
Carl Lange
Carl Lange
@halirutan Makes sense to do a meta post.
 
C. E.
C. E.
6:30 PM
It's interesting that Cuboid does not mention Cube in its Properties & Relations and Cube does not even have a Properties & Relations. Does anyone know what the point of Cube is?
Properties & Relations would have been good for Cube also because they could have explained the relationship between the arguments and Rotate/Translate...
 
b3m2a1
b3m2a1
6:44 PM
@CarlLange once I finally have the time to write a blog post about the whole tutorial plugin stuff I've been working I'll get around to releasing it so others can use it.
@VitaliyKaurov at one point or another you needed a Markdown -> Notebook converter. Do you still?
I took my basic Markdown parser and generalized it so you can plugin whatever post-processor you want and then wrote up one for notebooks
 
Kuba
Kuba
6:57 PM
@halirutan yes, I agree with doing it via Meta post.
 
Vitaliy Kaurov
Vitaliy Kaurov
@b3m2a1 that would be very interesting to look at. do you have a write up about it?
Guys I got a downvote on this one - does anyone see why? I should delete the answer if I missed the point:
https://mathematica.stackexchange.com/a/196968/13
 
Chris K
Chris K
7:18 PM
I've never rollbacked a post, but this seems worthy:
2
Q: Using procedual programming

rreewI'm very new to Mathematica. I want to solve these equations : $$exp(ikl)= \prod_{\substack{m=1\\m \neq l}}\frac{ \lambda_{l}-\lambda_{m}}{\lambda_{l}-\lambda_{m}}$$

finite-difference-method
existing answers don't even make sense any more
should I just roll it back to revision 6 or maybe a mod is better / necessary?
@VitaliyKaurov The question isn't very clear, but I think you overlooked the "it should delete {John, Doe} only if there is a string which contains already both of them" part -- I assume if there is no "John Doe", then "John" and "Doe" shouldn't be deleted
 
Szabolcs
Szabolcs
@user21 Have you ever used Mathematica's own random number generator for a library accessed through LibraryLink? One could in principle access RandomInteger, RandomReal and RandomVariable[NormalDistribution[]] through ConnectLibraryFunction. I am not sure how bad the performance would be.
The R and Python interface of igraph hook directly into R's and Python's random number generators. igraphs' RNG framework allows defining methods to get a random integer, uniform real or reals from various distributions.
These functions return a single value by default. Thus there is at least one function call involved for any generated number anyway. Now the question is how big is the overhead of Mathematica's compiled RandomInteger, etc. when generating a single value.
 
Vitaliy Kaurov
Vitaliy Kaurov
8:03 PM
@ChrisK yep, you are right thank you! deleted my response
@Szabolcs if you have a chance to chat with me a minute on FB - i'd appreciate
 
Chris K
Chris K
8:19 PM
@VitaliyKaurov no prob! wasn't my downvote BTW :) :)
 
Vitaliy Kaurov
Vitaliy Kaurov
@ChrisK haha no worries did not even think that
 
b3m2a1
b3m2a1
@VitaliyKaurov Not quite yet but I'll get it out when I write up my next thing on authoring/IDE stuff
 
C. E.
C. E.
8:36 PM
@Szabolcs thank you for notifying me
 
Carl Lange
Carl Lange
Right on time for 12's nice support of MIDI with SoundNote, and its implementation of the GPT network, OpenAI released MuseNet, which is basically GPT but for MIDI
 
 
2 hours later…
Vitaliy Kaurov
Vitaliy Kaurov
10:11 PM
@CarlLange yeah musenet is making its rounds in the media, nice one. And we got:
...GPT
 
 
1 hour later…
Carl Lange
Carl Lange
11:38 PM
@VitaliyKaurov Yup! Sadly I have nowhere near the resources to train GPT with anything :( (and also, there's no example of how to train/fine-tune GPT, or nearly any NN on the repo)
I would love to take some of our blog posts at toughsoles.ie/blog and train it on them, then do text generation and see what it comes out with
I did really enjoy some of the stuff it came out with without that though. twitter.com/csl_/status/1121551367926161409
 

« first day (2654 days earlier)      last day (1840 days later) »