I am trying to find a way to add control key to automate entering [[ and ]]
But the answers here automating-esc-esc-formatting do not work on 11.2 on windows any more. I've been trying for 20 minutes and they have no effect.
Should one in this case open a new question and add link to the old on...
This makes no sense. Maple shows up as #50 on Latest tiobe programming languages index tiobe.com/tiobe-index/ but Mathematica is not even on the list. M is not even in the list of 100 languages. This can not be right. someone from WRI should complain to that site about this index.
It's showing how often users are mentioned in my (small) corpus of comments vs their rep. Log plotted, of course, so that @Mr.Wizard doesn't break the Callout collision detection by causing too much clumping.
Either my corpus had a long conversation with @J.M. or he engages in a lot of comment conversations.
Anyone have experience with running Mathematica on a cluster? I'm currently being told that it is eating up extra cores when it shouldn't be. I've started the script with LaunchKernels with the appropriate number of cores, but it has apparently it is still grabbing everything.
I am looking for the fastest way to achieve the following:
Given a Graph, retrieve its edge list in terms of vertex indices (not actual vertices).
For example, given
Graph[{a, b, c}, {a <-> b, a <-> c}]
I am looking to get the output
{{1,2}, {1,3}}
as a packed array. Alternatively the ...
I am looking for faster solutions than the one presented.
Test graphs: g1 = GridGraph[{250, 250}]; g2 = Graph[EdgeList[g1]];. Note that IndexGraph[g1] is fast but IndexGraph[g2] is very slow.
Requirements: 1. The edge ordering must be preserved. 2. Multigraphs should be handled well. 3. You can forget about mixed graphs (those with both directed and undirected edges).
Solutions which are very fast but work only is certain (detectable) sub-cases are of interest too. The main function will have sub-cases based on what method is applicable.
I don't want to re-post the question, and I am not sure if putting a bounty is appropriate given that I already accepted an answer.
If you have any ideas, please let me know! I wasn't able to find a universally working fast solution yet. What I use in IGraph/M is a mishmash of special cases based on whether the edge ordering is relevant, whether the graph is weighted, whether it is a multigraph, etc. It's messy and bug prone. It's done this way so I don't have to pay the performance penalty of the most general solution when not necessary.
Whatever API Wolfram Community is using needs a few updates... It refused to accept a space as a last-name, but it's happy to accept a zero-width space. Also the lag time is brutal.
@b3m2a1 If you're looking for a parametric density function (i.e., one that fits the data with relatively few parameters) that can be "used" outside of Mathematica, then finding one using the Pearson system is one way to go: en.wikipedia.org/wiki/Pearson_distribution. This requires estimates of the square of skewness (3rd moment) and kurtosis (4th moment).
@b3m2a1 But if you only need the "use" of the distribution within Mathematica and aren't comparing distributions at different points in time or for different subsets of the data (comments from statisticians vs. physicists vs. chemists vs. engineers), then I'd stick with the results from SmoothKernelDistribution.
With results from SmoothKernelDistribution you can obtain any of the standard summary statistics (mean, 95-th percentile, median, etc.) for describing the distribution.
@JimB thanks for all the info. My knowledge of statistics is very basic so the pointers are very helpful. I was trying to fit multiple distributions because I thought I could see different dynamics in the plot (e.g. from people hitting the comment-length cap vs people essentially saying "thanks +1") but I will keep all of this in mind the next time I play pretend data scientist.
@Szabolcs @Kuba I need to write some documentation for some parts of my main package on GitHub. Would you recommend I do it in Markdown and expect people to view the docs on the repo page or should I provide built-in documentation as tutorials?
@b3m2a1 I'm not entirely up to date with the amount of pain it is to create and maintain built-in docs. I once did it around ~V10 for a customer and tried lately for a side project but I didn't finish. But once I'm back on it I will probably try to do GitHub based docs. It was not fun and is even worse to create proper documentation. I'm not going to waste more time on this.
So I'm not the best one to ask but that is what I had to say :)
@Kuba okay. I find them to be about equivalent amounts of pain, but I think that's because I have a built-in system for doing the docs in Mathematica, rather than in Workbench.
I mostly wanted to know which you thought would be most useful for the people who use the systems
I am trying to find a way to add control key to automate entering [[ and ]]
But the answers here automating-esc-esc-formatting do not work on 11.2 on windows any more. I've been trying for 20 minutes and they have no effect.
Should one in this case open a new question and add link to the old on...
I am looking for the fastest way to achieve the following:
Given a Graph, retrieve its edge list in terms of vertex indices (not actual vertices).
For example, given
Graph[{a, b, c}, {a <-> b, a <-> c}]
I am looking to get the output
{{1,2}, {1,3}}
as a packed array. Alternatively the ...
