|
Does anyone know the name of and/or a non-brute force algorithm for this problem? It's similar to the knapsack problem and the bin packing problem. It might be the multiple knapsack problem? Given two lists A and B of equal length, with each item in each list being an N-tuple, find the ordering(s) for A such that as many numbers as possible in A are larger than or equal to the corresponding number in B. A tie counts as half of a win. Example: A: [(1, 2), (3, 4), (5, 6)] B: [(2, 3), (1, 5), (4, 6)] The best ordering of A here is either [(3, 4), (1, 2), (5, 6)] or [(3, 4), (5, 6), (1, 2)], both giving 4 points.
|
# ? Mar 27, 2018 07:47 |
|
|
# ? May 18, 2024 05:07 |
|
That sounds like an instance of the assignment problem.
|
# ? Mar 27, 2018 14:32 |
|
This is only slightly related: is there somewhere/a thread I can offer up a small programming job for cash? I need something doing with Javascript.
|
# ? Mar 29, 2018 08:59 |
|
Henry Black posted:This is only slightly related: is there somewhere/a thread I can offer up a small programming job for cash? I need something doing with Javascript. Try this: https://forums.somethingawful.com/showthread.php?threadid=3246449
|
# ? Mar 29, 2018 09:09 |
Post it on freelancer.com, there's no lower limit even if the task is very small and cheap.
|
|
# ? Mar 29, 2018 11:09 |
|
That thread is archived, SA-Mart might be a good choice. Post a thread there and link it here for more exposure
|
# ? Mar 29, 2018 18:09 |
|
Start a new thread here in COBOL.
|
# ? Mar 29, 2018 18:11 |
|
Not really programming per se, but I couldn't find a better place. With Sublime Text 3, is there a way to disable the lorem ipsum text completion?
|
# ? Apr 1, 2018 13:22 |
|
I have a question about basic operations and algorithm order of growth. So this is my algorithm that I have to test: I would assume that the basic operation here is the comparison at the start, and because it's done exactly once for each index of the array, that the order of growth would be N. I decided to test it to be sure and I found that it isn't Am I missing something here?
|
# ? Apr 1, 2018 15:56 |
|
underage at the vape shop posted:I have a question about basic operations and algorithm order of growth. Dunno what to tell you. Possibly you're benefiting from some kind of runtime optimizations? What language are you using? Python code:
output posted:100: 0.0001s
|
# ? Apr 1, 2018 16:25 |
|
That is an asympotically linear growth curve. Its understood that wall-clock measurements wont necessarily match the curve for small N, for all sorts of fixed-overhead reasons like the time spent waiting for the code to load from disk. Your argument probably needs to be stronger than once for every element in the array, though, since its not like the code is obviously advancing through the array one element at a time.
|
# ? Apr 1, 2018 16:57 |
|
hooah posted:Not really programming per se, but I couldn't find a better place. With Sublime Text 3, is there a way to disable the lorem ipsum text completion? Have you tried setting auto_complete to false?
|
# ? Apr 2, 2018 13:48 |
|
Munkeymon posted:Have you tried setting auto_complete to false? No, because I want auto-complete.
|
# ? Apr 3, 2018 03:57 |
|
Destroy all software is free this week and subs are discounted. Has anyone here watched them? Would you recommend watching them? https://www.destroyallsoftware.com/sale
|
# ? Apr 3, 2018 18:40 |
|
I've watched a handful or two, theyre great. Youre watching someone good do their work, so youll learn something (or be reminded of something youve forgotten) even if the exact code on the screen isnt particularly novel or relevant to you. Hes good on twitter too, if thats something you partake in.
|
# ? Apr 4, 2018 03:19 |
|
I'm interested in creating a kind of diy plotter, basically a machine that uses a pen to draw shapes or images. What language and dev kits would I need to learn/acquire to make that possible? I'm new to coding and only got a year experience in Lisp, but I'm getting curious about C so maybe that would be a good start? I heard C is used in robotics, that's why. e: don't really care if there's a (cheap) machine out there that lets me do this, this is more of a learning project.
|
# ? Apr 4, 2018 09:42 |
Do you want to focus more on the software or on the hardware? How much do you know about electronics already? The "easy way out" would probably be a Raspberry Pi, install a Linux on it, and control stepper motors via the GPIO. You can do that in mostly any language you want, but C or Rust as low-level languages would give you the best timing guarantees. The somewhat more involved way would be with an Arduino kit (or a clone), in which case you'll almost definitely be writing C or assembly. You will probably also need to work out how to get the plotting data over to the device, since you don't have much working memory on those, it's almost real embedded development. But ask over in the DIY forum, they know more about it. Here's some suggested threads: https://forums.somethingawful.com/showthread.php?threadid=3505424 Arduino thread https://forums.somethingawful.com/showthread.php?threadid=3365193 3D printers thread (a plotter it almost the same, just with fewer motors) https://forums.somethingawful.com/showthread.php?threadid=2734977 Learning electronics megathread https://forums.somethingawful.com/showthread.php?threadid=3500975 Embedded development (here in CoC) nielsm fucked around with this message at 11:42 on Apr 4, 2018 |
|
# ? Apr 4, 2018 11:37 |
|
nielsm posted:Do you want to focus more on the software or on the hardware? How much do you know about electronics already? Next to nothing. I guess focusing on the software-side is enough for the start. I assumed it's an easy project for a first-timer. nielsm posted:The "easy way out" would probably be a Raspberry Pi, install a Linux on it, and control stepper motors via the GPIO. You can do that in mostly any language you want, but C or Rust as low-level languages would give you the best timing guarantees. Yeah, my first thought was Arduino and C. But that comment with the plotting data and the limited memory is interesting, guess I'll have to consider that. Will definitely check the other threads.
|
# ? Apr 4, 2018 13:42 |
|
Currently at work we're using a bespoke solution to start/stop and watch processes that are running on several different machines. This is all done via a web interface (and by watch, I mean we can see what's getting put out on stdout/stderr for that application) that sits on one of the computers. There's enough things we want changed, I've been trying to see if there are already any existing products in this space. Ideally it'd be able to run and manage processes on Windows as well as Linux (our current solution only supports Linux). We're using this to manage applications that have been written in Python, C++, and Node. The closest I've found is something like pm2 but it seems aimed only for Node, doesn't have a way to look at a programs stdout/stderr and I don't need anything to do with load balancing.
|
# ? Apr 4, 2018 13:57 |
|
mike12345 posted:I'm interested in creating a kind of diy plotter, basically a machine that uses a pen to draw shapes or images. What language and dev kits would I need to learn/acquire to make that possible? I'm new to coding and only got a year experience in Lisp, but I'm getting curious about C so maybe that would be a good start? I heard C is used in robotics, that's why. Comedy option: make a real-life LOGO turtle.
|
# ? Apr 4, 2018 16:51 |
|
I've been wanting to get more into ML stuff for a long time but haven't ever had a project to spur me into getting the learning done. Well now I think I have one. For a nerdy home automation thingy I'm wanting to use some raspberry pis with camera modules to recognize when I'm in different rooms of my house. Can recognizing specific faces (not just faces in general) in video be done on a pi in real time? I guess another option is I can send video from the cameras to my i5-based home server. Or maybe if the pi is good enough to recognize faces in general it can send snapshots of recognized faces to my home server for recognition...
|
# ? Apr 4, 2018 16:59 |
|
Youd probably want to train the network on your PC, but the classifier will probably run on a Pi. Edit: saw this the other day. Not exactly what youre doing but should inform what the Pi is capable of https://medium.com/nanonets/how-to-easily-detect-objects-with-deep-learning-on-raspberrypi-225f29635c74 Star War Sex Parrot fucked around with this message at 17:28 on Apr 4, 2018 |
# ? Apr 4, 2018 17:25 |
|
Since the functional programming thread is dead and archived, did anyone here go from Haskell to Purescript? They seem similar enough that it will be easy but I'm a little woried about the lack of laziness biting me.
|
# ? Apr 4, 2018 19:05 |
|
I didn't go from Haskell to Purescript, but I did go from Haskell to Idris (which is eager) and I didn't really notice any gotchas.
|
# ? Apr 5, 2018 01:39 |
|
TooMuchAbstraction posted:Comedy option: make a real-life LOGO turtle. Yeah, someone already did this.
|
# ? Apr 5, 2018 11:17 |
|
TooMuchAbstraction posted:Comedy option: make a real-life LOGO turtle. Logo literally started off with a real-life turtle, of course, that thing wasn't a metaphor. I'm not sure how useful learning sort-of-FORTH would be for today's embedded world, mind you, it's not 1975 any more.
|
# ? Apr 5, 2018 11:45 |
|
The 2018-appropriate version would be using GraphQL to talk to an IOT device that controls a living turtle via brain stimulation edit: this is what happens when you watch Black Mirror mystes fucked around with this message at 16:12 on Apr 5, 2018 |
# ? Apr 5, 2018 16:10 |
|
I'm going to be in the position of needing to redesign a major software platform at work and I'm lacking in some understanding regarding how well designed real time systems work behind the scenes. We have an application that currently just uses a MSSQL Server database as it's backend, the desktop application itself keeps an in memory snapshot of it's current state, and then just has some really poorly written HasChanged() methods to look at the state of the database compared to its cache to determine if another client changed it within X amount of time. It's a horrible mess of race conditions and collisions and requires a lot of supervision. I've been looking at various messaging and queue systems (RabbitMQ, Kafka, 0MQ, etc.) and I get the feeling that they were designed for use cases like this. Are there any pretty standard references or design patterns in regards to how systems like this are generally setup? Like how should the communication pathways be designed at the high level, how is long term storage handled, how to work with race conditions, etc. would be very helpful.
|
# ? Apr 5, 2018 16:20 |
|
feedmegin posted:Logo literally started off with a real-life turtle, of course, that thing wasn't a metaphor. I'm not sure how useful learning sort-of-FORTH would be for today's embedded world, mind you, it's not 1975 any more. Wait, really? That's not what I've read. Mindstorms mentions that the LOGO turtle was designed as such so kids could anthropomorphise it and "pretend they are the turtle" to solve problems, but never that it was a physical device.
|
# ? Apr 5, 2018 16:54 |
|
Suspicious Dish posted:Wait, really? That's not what I've read. Mindstorms mentions that the LOGO turtle was designed as such so kids could anthropomorphise it and "pretend they are the turtle" to solve problems, but never that it was a physical device.
|
# ? Apr 5, 2018 17:21 |
|
Suspicious Dish posted:Wait, really? That's not what I've read. Mindstorms mentions that the LOGO turtle was designed as such so kids could anthropomorphise it and "pretend they are the turtle" to solve problems, but never that it was a physical device.
|
# ? Apr 5, 2018 17:32 |
|
Suspicious Dish posted:Wait, really? That's not what I've read. Mindstorms mentions that the LOGO turtle was designed as such so kids could anthropomorphise it and "pretend they are the turtle" to solve problems, but never that it was a physical device. This post makes me feel old. vv- e: multiple people making the same mistake The Fool fucked around with this message at 17:41 on Apr 5, 2018 |
# ? Apr 5, 2018 17:33 |
|
peepsalot posted:LEGO didn't invent LOGO Mindstorms is a book by the logo dude.
|
# ? Apr 5, 2018 17:37 |
|
Master_Odin posted:solution to start/stop and watch processes that are running on several different machines. [...] we can see what's getting put out on stdout/stderr for that application [...] Ideally it'd be able to run and manage processes on Windows as well as Linux [...] I don't need anything to do with load balancing. Have you considered using something designed for botnets? I am only half joking. edit: I guess in generall, your search keyword would be orchestration, but I'm afraid that's completely overloaded with cloud and virtualization bullshit now. double riveting fucked around with this message at 18:30 on Apr 5, 2018 |
# ? Apr 5, 2018 18:26 |
|
Thermopyle posted:I've been wanting to get more into ML stuff for a long time but haven't ever had a project to spur me into getting the learning done. Well now I think I have one. Ghost of Reagan Past fucked around with this message at 02:26 on Apr 6, 2018 |
# ? Apr 6, 2018 02:23 |
|
Ghost of Reagan Past posted:The flow I would go for is that you have a face detection system on the Pi that takes pictures of every face visible to it (on an interval, maybe using OpenCV facial detection?), and pitches the picture to your home server for running through a trained model. I suggest using scikit-learn for your task. A Pi is probably powerful to do both, but I prefer this kind of workflow because it separates the two problems into two discrete tasks. I would in any event write it as two applications generally, the recognition application running as an HTTP server (likely with Flask), even if it runs on the Pi as well. The issue is that you're going to do a bunch of image processing, which can be complicated and computationally intensive. Cool, this is very similar to what I've got sketched out now. It's good to hear someone who sounds like they know what they're talking about saying that maybe I'm not barking up the wrong tree. Which reminds me...I too-often forget to say thanks when someone helps me because the thread has moved on to something else by the time my bookmarks list tells me there's new posts, so: Star War Sex Parrot posted:Youd probably want to train the network on your PC, but the classifier will probably run on a Pi. Thanks! That got me on the right path. Well maybe not, we'll see, but it feels like it helped!
|
# ? Apr 6, 2018 05:21 |
|
hooah posted:No, because I want auto-complete. What do you mean by "lorem ipsum autocomplete", then? feedmegin posted:Logo literally started off with a real-life turtle, of course, that thing wasn't a metaphor. I'm not sure how useful learning sort-of-FORTH would be for today's embedded world, mind you, it's not 1975 any more. LISP, not Forth.
|
# ? Apr 6, 2018 14:29 |
|
ToxicFrog posted:What do you mean by "lorem ipsum autocomplete", then? I mean when I type "l", I want the auto-complete to suggest a word I've actually typed, instead of a block of dummy text.
|
# ? Apr 6, 2018 16:08 |
|
I use ST3 on the daily and have never seen lorem ipsum come up in completion. Just tried with a few different syntaxes and a blank new file and it doesn't come up. What syntax do you have set? You sure it's not some installed package?
|
# ? Apr 6, 2018 19:07 |
|
|
# ? May 18, 2024 05:07 |
|
Ranzear posted:I use ST3 on the daily and have never seen lorem ipsum come up in completion. Just tried with a few different syntaxes and a blank new file and it doesn't come up. What syntax do you have set? You sure it's not some installed package?
|
# ? Apr 6, 2018 19:21 |