/dpt/ - Daily Programming Thread

1st for never use unsigned types unless it's about bit manipulation.

Oh godammit right as I posted my question, anyways can someone help me?

>>57179315

>>57179319
Ok... thanks for the advice...

>>57179319
This wouldn't happen for 64bit unsinged integers.

In openGL how do I rotate everything relative to the global origin?

I tried rotating my view matrix, but I lose all reference points or it breaks my game when I take the inverse

>>57179344
But then you have to store 64 bit integers which trashes your cache.

>>57179344
>unsinged
I guess they are just mute.
badum tss

>>57179319
>tfw Ada has unsigned integers with undefined overflows

>>57179350
Use another matrix purely for scale/transformation/rotation of a single object. Then you can rotate and move each object independently, as they all start out from the same global origin.

>>57179391
Based.

just to be sure, how many bits can a x86 memory adress hold, should be 32 bits right ?

13th for size_t will be a signed type in C++23.

>>57179300

int Message::getAgValue(unsigned int index)
{
    messageAg* ag = &m_ag;
    for (unsigned int i = 0; i < index; i++)
    {
        if (ag->nextAg == nullptr || ag->nextAg == DEL_MESS_PTR) return AG_MISSING;
        ag = ag->nextAg;
    }
    return ag->arg;
}

What kind of jobs can one expect to get out of learning assembly language?

>>57179448
blowjobs

>>57179448
300k starting

>>57179448
steve

>>57179398

Right now I have a uniform view and model matrix because I'm only rendering one polygon from one buffer

I just don't get how to make the model matrix have it rotate in local coords while the view matrix moves it around the camera (so rotates about the origin in global coords)

I get that's how you are supposed to do it though

I need help with simple c++ code
I need to convert ASCII lower case to upercase.
I don't know how to make this compile.

#include <iostream>
using namespace std;

char zamienNaDuza ( char litera )
{
litera = 'a' - 32;
return litera;
}

void wyswietlLitere ()
{
char litera = 'a';
cout << zamienNaDuza ( litera );
}

>>57179484
To backdoor iPhones fer dem government bois amirite?

> tfw my brother tries to learn Haskell as his first language
Does anyone know any beginner friendly, elaborate books for this dumdum?

>>57179500
Just multiply all the matrices in the right order. Apply the unique transformation matrix to your object vertices first, then the projection matrix, then the camera matrix. Then you do that with every object, each having it's own transformation matrix.

>>57179534
Learn you a Haskell for great good.
But the writing style is not for everyone.

>>57179391
They wrap around

>>57179547

So the order matters when you multiply them?

>>57179529
>'a' - 32;
Just include <cctype>

>zamienNaDuza
What does this mean?

And where's your main function?

>>57179534
Tell him to learn C. Haskell will only poison his soft and delicate mind, that is not yet accustomed with the intricacies of computers, with mathematical formulas.

>>57179568
I'm not talking about modular integers.
type Modular is mod 2**32;

I'm talking about regular integers.
type Unsigned is range 0 .. 2**32 - 1;
for Usigned'Size use 32;

Of course you have to disable overflow checking as well.

>>57179576
>So the order matters when you multiply them?
Yes. Immensely.

>tfw can't bring myself to open an editor and start programming

I don't get it, it's pretty fun when I actually do it. Instead I just go back to browsing memes or something... What's wrong with me and how do I stop it?

>>57179576
Ok, forget my post. Learn linear algebra first.

I started doing CodeEval challenges to improve at programming. I've just finished the sum of primes below 1000.
My solution was:

import math

# Get 1000 primes using the sieve of eratosthenes
def sieve():
    N = 89**2
    nums = [True] * N
    for i in range(2,int(math.sqrt(N))):
        if nums[i]:
            for j in range(i*i,N,i):
                nums[j] = False
    return [i for i,x in enumerate(nums) if x ][:1:-1]

# Sum the primes
sum(sieve())

The scoring system showed it used 6158918 B of memory and 113 ms to execute. I was wondering if there are ways to make this program using less memory/run faster in Python.

>>57179611
Have your editor run itself on startup, so it's already open.

>>57179608

wait, how come you do the projection matrix first then? wouldn't it be last since it moves shit into normalized coordinates?

>>57179300
how do i leave a process running during a ssh session after i disconnect?

so i want to ssh somewhere, run a process, then turn off my machine. i tried $ nohup run_script.py & but the nohup.out seems empty

>>57179611
Did you try not shitposting in programming threads?

>>57179639
Tried it, I just stare at the code for a while and then memes.

>>57179650
Yes, it's what I do most of the time.

It is my homework. I need to convert the bad script.

This is the script you need to edit to get uppercase letters (to iostream adds)

#include <iostream>
using namespace std;

char zamienNaDuza ( char litera )
{
return
}

void wyswietlLitere ()
{
char litera = 'a';
cout << zamienNaDuza ( litera );
}

>>57179664
>>>/adv/ fuck off

>>57179673
Step 1: Git gud.
Step 2: man ascii

>>57179673
>homework
>not in English
>not even formatted properly

Fuck off

>>57179636
>using less memory
By not allocating such a large array.

>>57179529
there is only one way to do this

char toUppercase(char c)
{
    switch(c)
    {
        case 'q': return 'Q;
        case 'w': return 'W';
        case 'e': return 'E';
        case 'r': return 'R';
        case 't': return 'T';
        case 'y': return 'Y';
        case 'u': return 'U';
        case 'i': return 'I';
        case 'o': return 'O';
        case 'p': return 'P';
        case 'a': return 'A';
        case 's': return 'S';
        case 'd': return 'D';
        case 'f': return 'F';
        case 'g': return 'G';
        case 'h': return 'H';
        case 'j': return 'J';
        case 'k': return 'K';
        case 'l': return 'L';
        case 'z': return 'Z';
        case 'x': return 'X';
        case 'c': return 'C';
        case 'v': return 'V';
        case 'b': return 'B';
        case 'n': return 'N';
        case 'm': return 'M';
        default: return '\0';
    }
}

>>57179673
>>57179529
>please do my homework for me or I'll keep spamming

Fuck off Polack.

>>57179597
sorry, can't do, his college demands it for his first class.

>>57179552
Sounds good, he is learning real world haskell

>>57179713
>>57179713

>>57179547

I had been doing them in the right order

Why does this (in my main loop)

        glm::mat4 view = glm::lookAt(
            glm::vec3(1.2f, 1.2f, 1.2f), // the position of the camera
            glm::vec3(0.0f, 0.0f, 0.0f), // the point to be centered on-screen
            glm::vec3(0.0f, 0.0f, 1.0f) // the up axis.  XY plane is the "ground".
        );

        view = glm::rotate(view,
            cum_x/200.0f,
            upAxis
        );
        view = glm::rotate(view,
            cum_y/200.0f,
            sideAxis
        );

where view is a uniform matrix, cum_x and cum_y is a mouse input, and my shader is this

gl_Position = proj * view * model * vec4(position, 1.0);

cause my only polygon to rotate in local axis?

>>57179713
>I made my doctor in computer science

>>57179598
I don't think the standards committee agree that a number that has the range of C's unsigned int is equivalent to C's unsigned int. For example, the interfaces packages defines unsigned_n as mod 2**n.

>>57179713
There is a standard library function for that, idiot.

>>57179719
I'm not sure Haskell as a first language is a good idea.
How does he even know about Haskell?

>>57179713
Post your diploma.

>>57179713
I'm annoyed you didn't leave an error in there in case he tries to copy-paste.

>>57179745
>#include <ctype.h>
>ever

File: 93u298j891.jpg (35KB, 239x600px) Image search: [Google]

35KB, 239x600px

>>57179713

>>57179756
Returning null bytes is already pretty bad.

>>57179734
Why is your camera a 3d vector? It should be a matrix. Read up on linear algebra first. Then a simple opengl tutorial.

>>57179738
>>57179745
>>57179747

Guys, I have some bad news for you...

>>57179746
The college enforces it to the fresh undergraduates in their algorithm and datastructure modules.

>>57179782
Don't tell me you didn't get 300k starting.

>>57179756
unfortunately i do
>case 'q': return 'Q;

>>57179713
This is the fastest way to do it if you have a decent editor like vim. It is also crystal clear what it does and what the intent is. Therefore it's the most pragmatic way to write it. Prove me wrong.

>>57179800
Yeah but that won't compile. I mean a subtle error like switching two letters so that his code fails any automated test the lecturer has set up.

>>57179808
>Prove me wrong.

char to_upper(char c)
{
    return c & ~0x20;
}

>>57179576
think of translating something up and then rotating it counter clockwise. that shoots up like a rocket, and then spins around the origin like a clock hand. if you rotated it 45 degrees, it'd end up far away from the origin at 3 oclock on the clock. if you rotated it 45 degrees and then translated it up, it'd lay on its side, and then shoot up to the 12 oclock. they end in completely different positions

>>57179778

> he didn't notice it was actually a matrix

How about you read up on linear algebra, brochacho

>>57179797
No, 320k

>>57179434

>>57179831
>bit magic
>some negated magic number or something
>not even base 10
>can't tell what's going on
>doesn't even work for unsigned chars

Nice try anon.

>>57179859
https://en.wikipedia.org/wiki/Fold_%28higher-order_function%29#Folds_in_various_languages

>>57179808

def to_uppercase(string):
     return string.upper()

>>57179921
>string argument
Doesn't adhere to spec, all tests failed, 0/10, see me next semester.

>>57179713
>default: return '\0';

You're a retard.

>>57179873

You too.
ASCII has been designed to work that way.

>>57179966
How can ascii have been designed if Design Patterns (TM) was first published in 1994?

>>57179966
Subtracting 32 would be way more obvious than masking the relevant bit of.

File: 1363902520808.jpg (38KB, 354x417px) Image search: [Google]

38KB, 354x417px

>>57179966
>not returning null character

File: himegoto C.png (1MB, 1000x1400px) Image search: [Google]

1MB, 1000x1400px

What's a good way to learn event driven programming in C?

>>57180007
>subtracting 32
nice UB

>>57180022
programing microcontroller

>>57180022
>event driven programming

But why?

>>57180022
It's just callbacks, big deal. Learn how qsort works and you're done.

>>57180022
man epoll

>>57180022
For starters, you need to stop being a literal degenerate.
After that, you can focus on programming.

File: talkshit.gif (2MB, 500x227px) Image search: [Google]

2MB, 500x227px

>wrote catch-all for a database query because time is short
>it sits on the server for a day or so, failing silently
>mfw I knew writing catch-alls were bad but I did it anyway

>>57179673
Did it on my own.
Thank you for the shitstorm!
#include <iostream>
using namespace std;

char zamienNaDuza ( char litera )
{
char duza;
duza = litera - 32;
return duza;
}

void wyswietlLitere ()
{
char litera = 'a';
cout << zamienNaDuza ( litera );
}

>>57180078
Do you want a medal or something? It was your homework to begin with.

>>57180078
>char litera = 'a';
>cout << zamienNaDuza ( litera );

just write

cout << zamienNaDuza ('a');

>>57180034
What's a GUI

>>57180127
A finite state machine.

>>57179319
>Negative array indexes.
Ok

>>57180078
>using namespace std;
not
>std::

What sort of degree would go best with CS to make me not be homeless?

Math/Applied Math? Statistics?

>>57180037
>>57180046
Thank you friendos.

>>57180175
Statistics lends itself to picking up machine learning which is what real developers are working on

>>57179529
>non-english identifiers
No. Just no. Don't do that.

>>57180200
>what real developers are working on
god i hope that isn't true
so far i can pick this stuff up but it's not fun

File: firefox_2016-10-22_00-17-06.png (77KB, 1564x791px) Image search: [Google]

77KB, 1564x791px

>tfw icc shits on your overflow
>tfw the generated assembly is twice as long as what gcc generates for signed integers
>but it (presumably) runs faster anyway
based Intel

>>57179344
>But then you have to store 64 bit integers which trashes your cache
you don't, simply promote before using them as array indexes.

>>57179319
that's wrong, there are other optimizations not possible with signed types.

>>57179407
32 on x86_32(32 bit) and x32(32 bit, but also kinda 64 bit)
64 bits on x86_64(64 bit)
what x32 is: https://en.wikipedia.org/wiki/X32_ABI

>>57179873
"ABCDEFGHIJKLMNOPQRSTUVWXYZ"[c-'a']
or c += 'A' - 'a';
both of which are faster than a switch table, and you should be used to tricks like these if you write C(++)

>>57180127
unnecessary bloat :^)

>>57180037
how it works in c, or in general
because I can understand haskell qsort perrfectly
but im not sure how its implemented in c

8 queens problem has 92 solutions and i can find them in 7 ms

if i really strain myself i can probably bring it down further

>>57180221
nice humble brag you cunt

machine learning is not difficult, calm down

>>57180256
i never claimed otherwise. i just don't like it. have you ever felt this way about something?

>>57180175
Anything else a programmer may end up dealing with. Off the top of my head:
>Math (usually for specialty areas like graphics): calculus, linear algebra, abstract algebra, discrete math, statistics
>Software Engineering
>Business stuff, management etc.
>Hardware & Electronics
>Domain-specific stuff, for example chemistry if you feel like writing molecular simulators
>Speaking of simulators, physics

That said, maybe you could even sell a sociology minor as an asset to an employer:
>My social studies allowed me to gain a deeper understanding of human nature and how great teams work together blablabla

Focus on something you're passionate about, so you can squeeze the most out of it through continued interest.

File: workinglight.png (24KB, 1472x520px) Image search: [Google]

24KB, 1472x520px

Great, I did it. Had to only apply the rotational matrix of each object to their normal vectors, not the transformation matrix.

>>57180286
I'm not really passionate about anything and I don't really have any hobbies or things I enjoy.


I'm expected to go to college so I went to college and just randomly decided on CS. I'm in my third semester and I figured it's easier to stick with it and try to make the best of it rather than start a different degree.

>>57180286
a math minor sounds like a good idea for him. i don't think you want to go too far unless you have the aptitude for it

>>57180236
In c, there's a function called qsort. It takes another function as a comparator. Not much else to it.

Word Ladders. Can someone explain how to organize word registry and building your tree from it?

>>57180318
Pick based on what you think you can do particularly well then.

>informatics 1
>have to write the EBNF for every programming assignment
dear lord, what kind of hell is this?

>>57180147
some languages support negative indices

>>57180320
I was talking about 2x BS degrees.

I feel like getting a job with just a CS degree will be hard.

>>57180139
Driven by user events.

File: 1452346762178.png (378KB, 1450x1080px) Image search: [Google]

378KB, 1450x1080px

What should I do as a programming project that would look good on a resume?


Would any of these work ?

>>57180397
where do you live
here in america, especially if you do some internships during the summers, a cs degree sounds great

>>57180377
Draw railway diagrams instead, then translate them to EBNF.

>>57180225
>"ABCDEFGHIJKLMNOPQRSTUVWXYZ"[c-'a']
>or c += 'A' - 'a';
Check an ASCII table before you post m8, you got the order wrong.

>>57180416
no, those are for the most part too small

>>57180397
If you apply yourself a CS degree is enough. Dabble with various languages and libraries in your free time, write personal projects, contribute to open-source, take on coding challenges etc. Have extracurricular activities that prove you're a good team worker etc.

>>57180379
Gross and illogical.

>>57179319
>1st for never use unsigned types
Opinion immediately discarded. Kill yourself immediately.

>>57179319
someone ask this on so
let's settle it

>>57180127

Garbage, as we all know.

>>57180497
It's fine if you don't think of it as a pointer offset.

>>57180497
>illogical.

hello retard

>>57180470
That's my problem with it.

It seems like it's hard to get a CS job when to get one you have to make your life about programming.

Why do I have to jump through all these hoops to sit in a cubicle and google shit and make 50k a year.

I don't want much out of life.

How to elegantly handle command line arguments in C? Let's say I have some optional arguments that also need a value, some optional flags that can only be used alone, some that can only be used with certain other flags. I feel like I always end up with a clusterfuck, I can make it work and have tried multiple ways but it always feels ugly as hell.

i didn't realize karlie was shilling for one of these coding bootcamps
dropped

a study was conducted

there are virtually NO heterosexual programmers

>>57180552

That's clearly untrue. Most programmer are normal. People on 4chan, in general, are not normal.

>>57180537
>50k
if you jump through the hoops you should be aiming to make much more than that

>>57180537
Do something else then, sell cars or start your own flower growing business or something.

>>57180542
https://www.gnu.org/software/libc/manual/html_node/Parsing-Program-Arguments.html

>>57180573
Idk, I don't really want a whole lot out of life.

If I could have a smallish apartment that isn't in a shit area, and be able to save a good amount of money, that's all I really want.

I'd rather work 30 or 40 hours a week and make less money than rake in the shekels but not have time to do anything but work.

I mean I guess that's a later career thing where you can get experience and then do consulting/work from home?

>>57180586
I don't really want to do anything. If I could not exist that would be fine with me.

>>57180573
suppose it depends where you live

>>57180599
>If I could not exist that would be fine with me.
Study Hume and convince yourself that you indeed do not exist.

>>57180542
Just use getopt.

File: cb_console_runner_2016-10-22_00-59-31.png (10KB, 942x125px) Image search: [Google]

10KB, 942x125px

>>57180439
how so?

>>57180466
>That's not all there is to it though.
>The use of sign/zero extending moves really makes it work.
how so? do you mean how it fetches the array elements into registers before adding them? i don't see how it helps deal with overflow

>>57180147
Perfectly valid.
Array indices are just pointer offsets after all.

>>57180845
It doesn't.
Icc either fails or is too smart and proved statically that they cannot overflow in this test.
I was a retard and deleted my post out of shame hoping nobody read it.

File: giphy.gif (2MB, 500x281px) Image search: [Google]

2MB, 500x281px

>2 years experience
>still suck

new career?

>>57180963
are you cute?

I recently finished my first Android app (IMHO, not a hacky buggy POS thrown together by pajeet) and published it on the Play Store. It has <1k total installs. I also released a library on GitHub that currently has >550 stars. I want to start applying for jobs, but I have no degree. What jobs and salary ranges can I hope to find?

>>57181003
sure

>>57181042
300k starting

>>57181129

R question

why does this works:

ggplot(data = pseudoFacebook, aes(x = friend_count)) + 
  geom_histogram(binwidth = 25) +
  scale_x_continuous(limits = c(0, 2000), breaks = seq(0, 2000, 25)) +
  facet_wrap(~gender)

but this doesn't:

ggplot(data = pseudoFacebook, aes(x = pseudoFacebook$friend_count)) + 
  geom_histogram(binwidth = 25) +
  scale_x_continuous(limits = c(0, 2000), breaks = seq(0, 2000, 25)) +
  facet_wrap(~gender)

The only difference is that in the second snippet I'm using the variable's "full name" (pseudoFacebook$friend_count), which I would assume would be easier to type since the intelligent typing thing from the IDE shows me the list of appropriate choices when I'm halway through typing the dataFrame name.

tfw u switch to clang from g++

feels clean man
feels fresh

>>57181042
Mind if I ask what kind of library it is? I've been thinking about farming stars on github for giggles. Find some new meme language and write whatever utility libraries for it that people are too lazy to write themselves, like left-pad.

>>57181203
>https://github.com/stevemao/left-pad/blob/master/index.js
is this a joke

>>57181194
Yeah dude. More readable errors, can be even more strict and a better manpage. From what I've heard gcc is more efficient but for someone learning this is very nice.

File: 1476970658209.png (128KB, 300x300px) Image search: [Google]

128KB, 300x300px

has anyone solved this one yet?
need a hand with it.

https://www.hackerrank.com/challenges/clique

>>57181220
It gets better http://www.theregister.co.uk/2016/03/23/npm_left_pad_chaos/

This is the current state of web dev.

Er det mulig å lære C++?

Can someone help me with my java homework? I'm retarded and can't do it

>>57181361
no

>>57181367
seriously, its due in a few hours, please

>>57181404
Just post it already

>>57181404
Can't help if we don't know what it is

>>57181237
i cant understand this site

is it finding a clique in a graph?

>>57181429
Det er lett å forstå. Er du dum?

>>57181349
英語以外の言語で書くことは可能ですか?

>>57181454
>>57181471
Svedupellehomot menkää vittuun mun laudalta perkele.

>>57181203
UI components

>>57181404
>>57181361
post it already i'm bored

File: tom.jpg (150KB, 500x642px) Image search: [Google]

150KB, 500x642px

Hey
I have question, I would like to write simple program that will download vid from yt.

I do not have too much experience with streaming media.
Probably I could google all that stuff with source code for application.

but lets assume that I would like to figure on my own what protocol is yt using for streaming media and how to use it in language.

How would you figure out what kind of protocol is yt using?

Is it possible from nmap/nc or something similar?

I want to create a basic hash table with residue method.

So far, I know that I can fill an array with n random numbers

int num;                                 
    srand (time(NULL));                                              
    std::cout << "Insert the array large: ";
    std::cin >> num;                                                       

    int *arrayPos = new int[num];                           
    for (int counter =1 ; counter <= num ; counter++)
    {
      arrayPos[counter] = (1 + rand() % 32000000);          
    }

And then, I could add the hash function (by residue method) as follows

int *arrayHash = new int[num];  
    for (int counter =1 ; counter <= num ; counter++)
     {
        arrayPos[counter] = (arrayPos[counter] % num);          
     }

After that, I don't know how to deal with collisions.

>>57181496
더러운 일본인 간다

File: goddess.jpg (180KB, 2000x1837px) Image search: [Google]

180KB, 2000x1837px

>>57181618
some are okay

>>57181634
>Korean "women"
https://www.youtube.com/watch?v=FiJd_d8I-iE

Is there any reason to use C over C++ these days?

>>57181538
wireshark, but if you have to ask like that i doubt you can do it

>>57181659
Is there any reason to use C++ over C# these days?

>>57181669
Is there any reason to use C# over JavaScript these days?

How can I make a for loop in Python to calculate and output interest and interest+base amount for each year across 10 years? Shit's bugging me.

Is there any reason to use anything but Ruby, now that it's 2016 and not the 90s?

>>57181695
well memed

>>57181695
Yes there is

>>57181683
Is there any reason to use JavaScript over Common Lisp these days?

>>57181691
So you want to see a list of interest and the current balance for every year?

>>57181714
no

either use C or use python

rest is meaningless

maybe assembly

Is there any reason to use signed integers over unsigned ones these days?

How can I average two integers in lambda calculus?

>>57181739
Is there any reason to use types these days?

>>57181726
Yeah, I know how to print the outputs and stuff, but whenever I try doing it over a range(10) it just multiplies base+interest by 1, 2, 3, 4, etc.

>>57181750
is there any reason to do anything these days?

>>57181659
>Is there any reason to use C over C++ these days?

Not really. The only consideration is if you were thinking of porting to a platform that doesn't support C++, and those platforms are increasingly rare. (Example: Rabbit 2000.)

If you're conservative, you can use the C++ compiler and limit yourself only to C++ constructs that add no overhead. For example: namespaces, references, constructors, destructors, public, private, etc. don't add any overhead beyond the equivalent C code. New/delete speed is probably close to malloc/free speed, but definitely check it out of you're concerned.

>>57181756
no ㅜㅜ

>>57181756
This.


Why haven't you kys already?

>>57181756
no
i thought we established this a while ago

>>57181768
My brain doesn't let me

>>57181760
Thanks!

>>57181779
you are not me you fucking shit

>>57181429
finding the maximum possible clique give the number of nodes and edges.

>>57179414

Aside from the fact that this would break compatibility with a lot of existing code, there is also the fact that we already have a type to do just that: ptrdiff_t. Also ssize_t if we can assume an at least semi-posix system.

>>57181575

>how to deal with collisions
Hash tables are not flat arrays. They're usually implemented as arrays of buckets, where a bucket could be a vector or a balanced binary search tree of sorts. When you insert something into the table, you will hopefully be able to place it into an empty bucket, thus giving you O(1) time to insert or pull things out. From time to time you will have things that correspond to the same hash value. So insertion or finding something is going to take either O(n) or O(log n) time depending on how that bucket is implemented.

Ideally, you will also want to let the size of your table grow, and the hash function to mutate itself as you get more entries.

>>57181575
Guess you ain't helping

>>57181695

Yeah, C if you want to develop native extensions for Ruby.

File: firefox_2016-10-22_02-34-45.png (19KB, 304x59px) Image search: [Google]

19KB, 304x59px

>tfw captcha wants me to type 6000 spaces
>>57181760
>If you're conservative, you can use the C++ compiler and limit yourself only to C++ constructs that add no overhead.
or you could keep using C like most people

>he fell for the "C is outdated" meme
loling @ ur life fException in thread "Local computation of job 12" java.lang.OutOfMemoryError: Java heap space
at org.apache.spark.api.python.PythonRDD$$anon$1.read(PythonRDD.scala:131)
at org.apache.spark.api.python.PythonRDD$$anon$1.next(PythonRDD.scala:119)
at org.apache.spark.api.python.PythonRDD$$anon$1.next(PythonRDD.scala:112)
at scala.collection.Iterator$class.foreach(Iterator.scala:727)
at org.apache.spark.api.python.PythonRDD$$anon$1.foreach(PythonRDD.scala:112)
at scala.collection.generic.Growable$class.$plus$plus$eq(Growable.scala:48)
at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:103)
at scala.collection.mutable.ArrayBuffer.$plus$plus$eq(ArrayBuffer.scala:47)
at scala.collection.TraversableOnce$class.to(TraversableOnce.scala:273)
at org.apache.spark.api.python.PythonRDD$$anon$1.to(PythonRDD.scala:112)
at scala.collection.TraversableOnce$class.toBuffer(TraversableOnce.scala:265)
at org.apache.spark.api.python.PythonRDD$$anon$1.toBuffer(PythonRDD.scala:112)
at scala.collection.TraversableOnce$class.toArray(TraversableOnce.scala:252)
at org.apache.spark.api.python.PythonRDD$$anon$1.toArray(PythonRDD.scala:112)
at org.apache.spark.api.java.JavaRDDLike$$anonfun$1.apply(JavaRDDLike.scala:259)
at org.apache.spark.api.java.JavaRDDLike$$anonfun$1.apply(JavaRDDLike.scala:259)
at org.apache.spark.SparkContext$$anonfun$runJob$4.apply(SparkContext.scala:884)
at org.apache.spark.SparkContext$$anonfun$runJob$4.apply(SparkContext.scala:884)
at org.apache.spark.scheduler.DAGScheduler.runLocallyWithinThread(DAGScheduler.scala:681)
at org.apache.spark.scheduler.DAGScheduler$$anon$1.run(DAGScheduler.scala:666)

>>57181921
>most

>>57181945
>there only two languages are C and Java

File: Capture.png (79KB, 730x356px) Image search: [Google]

79KB, 730x356px

>>57181518
>>57181421
>>57181415

>>57181851
np complete

find all subsets of vertices and figure out if they are a clique, and if so then are they a maximum clique

i suppose you can use the maximum degree of a node to try and make it faster if the graph is sparse

it's exponential either way

File: Capture1.png (74KB, 649x780px) Image search: [Google]

74KB, 649x780px

>>57181518
>>57181415
>>57181421
>>57181982

>>57181982
so whats hard about it?

>>57181999
anon was being a little careless. it's asking instead for the largest clique that is guaranteed to exist given |V| and |E|

How do I start up a driver in C? Do I just fork and execl it?

Is there a way to check if there is already a process for that driver is currently running

>>57182035

import java.util.Scanner;
public class Assignment04 {
    public static void main(String[] args) {
    Scanner scan = new Scanner(System.in);
    //declaring variables for temperature conversion
    double f2k;
    double f2n;
    double f2c;
    double k2f;
    double k2n;
    double k2c;
    double c2f;
    double c2n;
    double c2k;
    //declaring variables for the start and end temp input, and a true/false for the correct temp unit input
    double startTemp;
    double endTemp;
    boolean correctTempUnit;
    String unit;
    do {
        System.out.println("Please enter a temperature unit: F, N, K, or C.");
        unit = scan.nextLine();
        //making all input uppercase, so later on I can analyze the text using just capitals
        unit = unit.toUpperCase();
        //correct units are f, n, k, c, and the user can exit
        correctTempUnit = unit.equals("F") || unit.equals("N") || unit.equals("K") || unit.equals("C") || unit.equals("EXIT");
        if (!correctTempUnit);
        System.out.printf("Invalid temperature unit.\n");
        else if (unit.equals("EXIT"));
            System.out.printf("Exiting program.\n");
            else {
                System.out.printf("Enter a starting temperature.\n");
            startTemp = scan.nextDouble();
            System.out.printf("Enter an ending temperature.\n");
            endTemp = scan.nextDouble(); }
    }
        while (startTemp <= endTemp) {
            
        }
    scan.close();

    
    }
}

obviously this doesn't work. i need to study but this homework's due in like in hour so i'd appreciate help.

File: turan.png (56KB, 1231x857px) Image search: [Google]

56KB, 1231x857px

>>57181999
>>57182037
Nevermind guys, managed to find the solution.
Here the code if anyone is interested.
Needless to say I couldn't have figure this shit by my own.

>>57181969
like most people who want to use C, yes. There are also other reasons, such as that interfacing with C++ code is a pain in the ass due to name mangling etc.

>>57182172
you could, you probably cant into graphs or something, it's not hard

not to say i understand this solution at all

>>57181739

My rule is to use unsigned, unless you need the signed feature.

For example, if a variable is used to index a C array, there is no possible way it could legally be negative, hence it should be unsigned.

But if you're ever subtracting things, then you might have a case where the result could legitimately be negative. So in that case use signed. An example of that would be pixel coordinates on the screen -- you might want to support allowing something to be placed off-screen by giving it negative pixel coordinates.

It all comes down to this: "Should it be illegal for this variable to ever be negative"? If yes, then make it unsigned, otherwise use signed.

FYI, in my 25 years of software engineering practice (mostly C/C++ on UNIX/Linux), I would say that about 80% of the time I used unsigned. By far my biggest usage of signed was for functions like the POSIX "int read()" function, where it returns -1 as an error code, hence you need signed. Another big case was getchar() that returns -1 on EOF -- the result must be signed int so that the -1 on EOF doesn't get confused with the character '\xFF'.

>>57182035
please help

>>57182166
>if (!correctTempUnit);
you ´should put a { } block behind if, like you did with the do while loop.

>>57182226
Thanks, this has been rather informative

File: masterrace.png (65KB, 997x678px) Image search: [Google]

65KB, 997x678px

>>57182166
well the first step would be to get this thing to run. java should be giving you really explicit messages here
you seem to be overdoing it with semicolons. after an if (condition) it's looking for a statement to execute if the condition is true. when you write if (condition); you end that statement immediately. the same thing happens after your else if

>>57182172
nice, i'll take a look when i have more time
do you like hackerrank? i'm overwhelmed by the number of sites like it. i don't want to compete, just practice

>>57181921
>or you could keep using C like most people

Or you could evaluate each C++ feature, and if it seems useful then use it.

A good example is constructors. Constructors add no overhead beyond the equivalent C code, but they can help you avoid programming errors by guaranteeing that variables are always initialized properly.

Of course you never NEED constructors -- but that's not the point. The point is to consider using a coding style that minimizes potential mistakes, and constructors can be a useful tool for that. If you're only writing a small amount of code, it doesn't really make much difference. But if you're part of a team of 10 developers all working on a million lines of code, then the benefits start adding up.

>>57182166
after you get your braces right it's pretty simple
>make methods converting temperatures (that's simple)
>make method that returns complete formated line (that's simple)
>you can make another method for outlining (simple)
>use above in loop, incrementing startTemp by one (simple)
>???
>you have at least 20 more minutes left

any cinnameg experts can tell me what's wrong with this?

Error:
  There is a syntax error that I cannot understand.
  Sometimes this is caused by an identifier of the wrong kind,
  such as a type identifier where a data item is expected.
  It can also be caused by an unknown type identifier.

-   34.   /\? : FormulaTester
          ^

  Recently read token: /\?

  Recent tokens:
    neg?                       [identifier]
    :                          [colon]
    FormulaTester              [type/genus/class identifier]
    /\?                        [identifier]

  Constructs currently inside (smallest first)
    line    32: Expect
    line     1: Package

  (Identifier "/\?" the cause?)


sequent.cmg:
There were errors.

/\? is the function name he told us to use. here's the whole expect block

Expect
  neg?: FormulaTester
  /\? : FormulaTester
  \/? : FormulaTester
  ==>? : FormulaTester
%Expect

>>57182331
>Or you could evaluate each C++ feature, and if it seems useful then use it.
look, people just don't do this in practice
why? that's a good question, there are certainly some reasons.

>>57182331
>but they can help you avoid programming errors by guaranteeing that variables are always initialized properly.
and what if i don't want to initialize a variable because i know i'm going to write to it?

File: cinnameg-masterrace.png (130KB, 760x1564px) Image search: [Google]

130KB, 760x1564px

>>57182379
please respond

File: 1427714516466.jpg (23KB, 600x541px) Image search: [Google]

23KB, 600x541px

>>57179713
>written in QWERTY order instead of alphabetic
>he actually typed the letters in there manually

you meme'd yourself kiddo.

>>57182160

Drivers/kernel modules are not processes.

https://linux.die.net/lkmpg/x44.html

>>57182222

for(var i = 0; i < input.length; i++){
        var arr = input[i].split(" ");
        var nodes = parseInt(arr[0]);
        var edges = parseInt(arr[1]);
        var r = nodes - 1;
        while(true){
            var m = (1/2)*(Math.pow(nodes,2)-(nodes%r)*(Math.pow(Math.ceil(nodes/r),2)) 
                    - (r - (nodes % r)) * (Math.pow(Math.floor(nodes/r),2)));
            if(edges > m){
                console.log(r + 1);
                break;   
            }else{
                r--;
            }
        }
    }

Basically is the application of the Turan's theorem.

            var m = (1/2)*(Math.pow(nodes,2) - (nodes % r)*(Math.pow(Math.ceil(nodes/r),2)) 
                    - (r - (nodes % r)) * (Math.pow(Math.floor(nodes/r),2)));

Calculates the amount of edges in a complete graph m given a number of nodes and edges, it starts by setting the largest possible subset

r = nodes - 1

and start iterating, we make sure that

edges > m

, if true it means that the current subset plus one node r + 1 is the largest clique subset possible (as in how many nodes it will contain), otherwise we just decrease the size of the subset r until we find a number that fits.
>>57182300
I've been there for about a week, I like that they have some areas focused exclusively in specific interests like learning programming languages, security, linux shell and such.
However I think some of their exercises are really misguiding, they either purposely don't mention a key factor that you will need or is poorly written leaving you with many doubts. They have discussions for every exercise and people seems to care enough to help there. They also give few test cases which makes can be good and bad depending on the situation. Codewars might be something more fit for what you want, poor language coverage in some cases, but it doesn't have the currency system.

File: halp.png (34KB, 1238x354px) Image search: [Google]

34KB, 1238x354px

>>57182674
forgot pic

>>57182627
sorry senpai, I guess it doesn't really classify as a driver.

Execing it seems to work fine, reading in from the socket seems to work too.

>>57179611
i feel you fjriend

>>57182681
ceiling and floor, i. e. round up and round down

>>57182707
Yeah, I figured that one out. [spoiler]Still had to ask on >>>/wsr/[/spoiler]

>>57179831

char to_upper(char c) {
  return c-1<3<<5|c>123?c:c&~' ';
}

>>57182681
this is iverson's notation
does anyone use apl anymore?

>>57181163

Anyone?

Why friend_count works but pseudoFacebook$friend_count doesn't? Aren't both calling for the same variable?

>>57182395
>look, people just don't do this in practice

Well, they don't do it in practice, until they do.

I've worked for about 10 companies (all C or C++ shops) and I was working all throughout the biggest C-to-C++ transition years (roughly 1995 to 2010). I've seen numerous C-only shops start investigating the advantages of C++ -- and then eventually over a few years, start using C++ more and more for new code as the developers see the advantages, or as new developers come on board. Some switched early, some switched later, and some never switched. I had to guess, I'd say roughly 3/4 of the C teams I've seen eventually ended up becoming C/C++ teams or else C++ only teams over time. It was usually triggered by some big change, like for example, starting to use Microsoft MFC to code the Windows GUI -- in that case, you had no choice but to bring C++ into your shop. And once developers start using C++, a pretty significant percentage of them wanted to continue to use it, and they resisted going back to C only. Eventually I even saw C++ make it way down into embedded firmware projects.

>>57182627
I guess its more of a Daemon, I need to check if its already running and if not I need to start it up. On further inspection exec isn't working for me

>>57182889

ps -e

Can someone post an example of a c program using anything that could be described as weird tricks? Preferably small.

>>57182965
yeah I need to do that programatically in C though

>>57183010
fast inverse sqrt is a classic

>>57183010
Duff's device comes to mind:

send(to, from, count)
register short *to, *from;
register count;
{
    register n = (count + 7) / 8;
    switch (count % 8) {
    case 0: do { *to = *from++;
    case 7:      *to = *from++;
    case 6:      *to = *from++;
    case 5:      *to = *from++;
    case 4:      *to = *from++;
    case 3:      *to = *from++;
    case 2:      *to = *from++;
    case 1:      *to = *from++;
            } while (--n > 0);
    }
}

>>57183010
GNU implementation of strlen()
http://www.stdlib.net/~colmmacc/strlen.c.html

File: voting.gif (83KB, 510x417px) Image search: [Google]

83KB, 510x417px

I made a simulator for Soros' voting machines.

Why is GHC so goddamn slow?

https://twitter.com/dhh/status/789635598323556352

crud man is at it again

>>57179529
kurwa

I've optimised the som code further, now it runs 4 times faster, but at the cost of memory.

I love low level.

>>57183245
waht a cuck

>all those people saying he should be kicked because "fascism is not REAL politics"
lo

>>57183338
l

>>57183204
why is it trump? the DNC doesn't need to do that to win, on the other hand they had to do that against bernie sanders those fucking crooks

>>57183461
>the DNC doesn't need to do that to win

Obviously they do.

File: questoon.jpg (494KB, 1372x1055px) Image search: [Google]

494KB, 1372x1055px

Should I continue to pursue a career in computer science if I have no passion for it and have no desire to do it in my spare time?

On the other hand, I'm fairly good at it and programming is currently the work that I have the most fun doing at my Uni.

>>57183493
>i don't like programming

stop right now, you're going to hate your life

>>57183471
oh I forgot she backstabbed millenials by going even more conservative for nationals after bamboozling bernie sanders and the american people

>gets the most conservative dem you can possibly ever have as VP
what the fuck

>>57183493
you won't make it unless programming, reading books and documentations and wanting to understand them is your hobby

>>57183534
I mean, don't most people not love doing their jobs?
Do I really need to love programming to work in the field?

File: eternity.jpg (86KB, 770x573px) Image search: [Google]

86KB, 770x573px

What's the name of this effect, where a material renders as a texture constantly billboarded to the camera as if it was a window?

installing gentoo.

>>57183541
>oh I forgot she backstabbed millenials by going even more conservative for nationals after bamboozling bernie sanders and the american people

Jesus, liberalism really is a mental illness, folks.

>>57183573
its fine that you don't have any passion for it, just aim for the money if you want

but trust me you will need more than just spare time and a degree on CS to make it
if you are not willing to put in that work, than choose law or something
it will get you more cash with less time input

just note that the time input will be 100x more boring than programming though imo

>>57183607
nothing constructive to say?

>>57183573
And this is why you're going to be blown the fuck out when you enter the working world, because you'll be competing with pajeets and autistic nerds who live breathe and eat software development and are unironically "passionate" about it.

File: artificial-intelligence-378x225.jpg (30KB, 378x225px) Image search: [Google]

30KB, 378x225px

Boss wants me to make an artificial intelligence program for classification. The features being looked at can be split into 2 broad categories; each broad category has many smaller features, all of which are binary boolean (yes/no) parameters.

My linear classification system treats each small feature as a feature to examine, giving be around 700 inputs to the system.

Right now, I'm sitting on a super shitty 55% accuracy using Tensorflow and a modified form of the linear classifier tutorial:
https://www.tensorflow.org/versions/r0.11/tutorials/wide/index.html
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/examples/learn/wide_n_deep_tutorial.py

What should I do to increase accuracy? Am I even using the right method?

Sorry for being very abstract, it's just that I cannot reveal what it is since it's for work, and I have literally done nothing relating to AI before.

>>57183651
>>57183610
Isn't there just some kind of code monkey job where I make a moderate amount of money doing monotonous work?
I don't want to be successful, I just want to get by.

The way the people at college talk the job market is just begging for any computer science graduate it can take. I mean, I know that isn't true but I'd hoped that I could get a low-level job.

>>57183651
tfw there was a time in life I was outside of /dpt/, reading docs, and gettingud
I miss fun, maybe
I don't know anymore, code is all I live for

>>57183153
That's fairly clever.

>>57183699
Honestly, there are better careers if your only goal is making money, and most of them take much less time to achieve proficiency.

>>57183699
just by the looks of your attitude there is a pajeet out there who will take the lowest you will want to get out of a CS degree and be 5 times more competent than you

given you don't put in the time in all your spare time

>>57183092

Not necessarily portable to other operating systems, but...

http://www.tldp.org/LDP/sag/html/proc-fs.html

Open, read, readdir, etc... Basically just start peeking at all of the numbered directories in /proc to see if any of them contain the name of your program.

>>57183625

Nothing you said was an argument. Sanders voters are just as bad as Clinton voters.

>>57183742
>>57183766
Fair enough, I guess.
Why does it seem like this is the only field where you need to be super dedicated and passionate to make it anywhere in it?
I feel like I've been bamboozled by every single person I know telling me computer science is a "safe" field to get into.

>>57183809
>2016
>thinks that 1% > 99% is logical
if this was a meritocracy on the other hand, and inheritance did not exists, I am down.

call me butthurt at the dice roll of life if you want though, but trust me it is a retarded system where engineers and scientists don't sit at the top even though they are literally the ones getting us forward

>>57183809
Sanders was objectively the correct choice for most people's actual value system, when you strip away all the tribal stuff and distortion of one's own desires. Comparing him to Clinton shows how little you actually know about anything.

-Applied crowd psychology
-Induced tribalism
-Divide and conquer

Whether ancient Rome, or today, it remains the time honored method of governance. You are not informed, you are just another product.

File: ayy.jpg (34KB, 600x403px) Image search: [Google]

34KB, 600x403px

Suppose you had internship offers from both Google and Facebook, and the compensation was the same. What would you choose?

>>57183881

>thinks that 1% > 99% is logical
It kind of is. You produce goods, you sell them for a profit. You gain money while they lose money. You keep reinvesting and selling, and your net worth keeps going up. Most people in the world aren't selling goods, but labor, which may be more reliable, but it isn't worth much because everyone can offer their labor. If you want to get up to 1% levels of money, maybe you should either acquire a skill or invent a product that plenty of people want to purchase.

>>57183936
google

>>57183936
Google.

>>57183936
depends on the office

>>57183936
Google. I'm more curious in their God than Facebook's. Facebook is just a retake on classical casino models. It's already been traced out how human social dynamics work and how to control people.

Google is doing interesting things.

>>57183853
>computer science
This isn't even a field anymore unless you intend to work as a data scientist with big data.
All the major accomplishments of CS stopped in the early 1970s, CS is a dead field.
The only reason anyone still takes it is because it has "computer" in the name and everyone thinks it's "personal computers" and not "information science/computation science".

Unlike most job paths, software development is one of the few careers you can study entirely on your own without certification or traditional schooling of any kind.
Everything you could possibly want to learn about software engineering is online, the only thing you need is time and motivation.

You would be stupid to go to college just to get a job in computers in 2016.
That job isn't even guaranteed anymore, software development is very saturated.
But even if you fail, at least you didn't start your working life with $50,000 of uncancellable debt before you've even gotten your first paycheck.

>>57183923
>Sanders was objectively the correct choice for most people's actual value system

You've already lost the plot here -- not everyone is a hyperliberal, nor do they believe that wanton spending is the solution to everything. There are a wide variety of value system that people in the US apply to their lives, and Sanders didn't have a monopoly on some objectively correct one.

>when you strip away all the tribal stuff and distortion of one's own desires

'Feel The Bern', 'Berniecrats', etc are the same expressions of tribalism within the culture.

There's nothing about Sanders that was unique.

>>57183881
>if this was a meritocracy on the other hand, and inheritance did not exists, I am down.

Why shouldn't inheritance exist? It's your money. What you do with it upon your death is up to you.

File: [DeadFish] Working!! - 07 [BD][720p][AAC].mp4_snapshot_07.11_[2014.10.02_22.17.58].jpg (48KB, 1280x720px) Image search: [Google]

48KB, 1280x720px

>>57183936
Google, just to be near with mootykins.
No homo.

>>57183945
Alright, suppose Facebook paid like 15% more. Would you still?

>>57183957
Menlo Park for Facebook, most likely Mountain View for Google (haven't host matched yet so not certain).

>>57183961
That's my fear - that I pick Facebook and end up doing CRUD work, or Google and host match to spud picking in Idaho.

>>57183967
Question that I know the answer to:

Is Information Science and Technology any better than Computer Science?

>>57183941

That's hard work, Ruby. We should just tax 100% of their income over 500k and we can spend that on things that help everybody else! They don't need all that money!

:^)

>>57183996
Keep getting fucked then, I guess. You can only try the same stupid thing in so many ways, over so many centuries, before you learn it ain't gonna work.

>>57184027

But socialism is tried and true, which is why Sanders' plans were sure to revitalize this ailing economy.

I can't think of a time it ever went wrong.

>>57183936
Google. They've actually got a couple of projects I wouldn't mind working on. I would love to do some core development on Android, for instance. Facebook... I feel as though it would be awkward to work on a product that I will never use willingly.

>>57184020
>That's hard work, Ruby.
Yeah, well, if you want a comfortable life, you should have to do hard work.

>>57183695
How many categories are you using? How much training data do you have?

>>57184064
>Yeah, well, if you want a comfortable life, you should have to do hard work.

I think that's what I implied. The sarcasm should have been obvious. I even included Mr. caret nose.

>>57183954
>>57184064
Would you still pick this if Facebook paid more? I kind of feel like I might be more likely to get more interesting work, but I also feel like I might be throwing away a huge opportunity whichever one I pick.

And Facebook's rated higher on glassdoor too...

>>57184043
Stop thinking in isms, and ists, and think mechanically. Actually understand how things presently work, and can work, so you can generate realistic predictions.

This is why I said you don't know anything. You don't. Which is why you never go in deep to how you actually engineer a society that works, and why our present one does not.

>>57183936
Google has better developer infrastructure. Ask any ex-Googler working at Facebook, and they'll tell you the same thing.

I'd choose Google for that reason, to see what a well oiled dev environment is like.

>>57184064
google has one main product that you're somewhat forced to use, as does facebook. the only difference is that google's was once good
they've both got interesting projects

>>57183695
Linear classification only works if your data is linearly separable.

Imagine this: you have 2 classes in a 3D space. Each class is represented by a ball-shaped cloud of data points. If you can draw a sphere around the centers of those balls where all points inside the sphere are class 1, and all outside are not class 1, and for the second sphere vice versa, then you can draw a plane into the space that separates them linearly. The more those clouds overlap, the less your classifier will work.

You might want to look into nonlinear classification if linear methods yields shitty results. Although I can't guarantee it'll get better. Many real world problems aren't solvable without either enough input vectors or enough abstraction & cleaning.

>>57184115
>>57184132
Yeah, I could imagine that. And Facebook really has one main product with different facets, whereas Google has many different products that integrate together.

>>57183941
Why are you taking off inheritance and system abuse out of the picture? Why are you talking about labor as something one dimensional and not about specialty, the most fundamental part of economics? There are more millionares in the world, then lets say a natural language processing specialists for AI or crisp biotech engineers who can actually benefit society
I don't see them get millions, even though I consider them to be worth beyond so

>>57184149
>Why are you taking off inheritance and system abuse out of the picture?
Because everything is a super simple concept when you take the difficult parts out of the picture.

>>57184020
meritocracy implies if you don't work your ass of you won''t get far in life

>>57184108
>Which is why you never go in deep to how you actually engineer a society that works, and why our present one does not.

You couldn't engineer your way out of a wet paper bag, you slimy fuck.

Get a job, bum.

>>57184173
>everything is a super simple concept when you take the difficult parts out of the picture
And everything is also a super convenient picture when you take the parts you don't like out

>>57184185
I had it with you, glad you are using a trip to simply filter you

>>57183577
people can't tell what you're talking about by that picture because it's a still image

>>57184198
whats getting taken out, tell me
inheritance and system abuse is in the picture

>>57184199

If you put as much effort into getting as job as you did being angry with me, you'd actually contribute to society.

Then you'd have no need to come on a Korean line dancing enthusiasts bulletin board to whine about meritocracies, bigger pictures, and abolishing the abusive system of people being able to do whatever they please with their own property.

>>57180963
I've watched this for 10 minutes. This is truly amazing.

>>57184185
That's what I thought.

Nothin' to see here. Just another scared, confused, angry know-nothin' care-not fuckhead. Isn't it ironic that you yourself are evidence of why people can't manage themselves? No matter what happens, there will be people like you. Either you're being controlled unwittingly, and in the way. Or you're delusional and too stupid to understand yourself and your environment, and are in the way. In either case I'd like to create a system where you're put to good use instead of ending up flailing around, like you are right now.

Collectivism and individualism can be balanced.

>>57184228
I have a job and trust me I contribute more to society than you.

>>57184240

I doubt that, commie.

File: Eternity2.jpg (96KB, 431x768px) Image search: [Google]

96KB, 431x768px

>>57184201
Right, sorry...
Effectively the model is operating as a "window" to a texture. here's a better view.
Sorry for gw2, but it's the best example I could find

>>57183577
>>57184270
what
are you talking about particles?

>>57184268
For what it's worth, I'd try to re-educate you before condemning you to the gulags.

>>57184270
As if she has a green sword and you greenscreen a static background?

>>57184294

No. He's saying that the 3D object doesn't actually have a static texture, but it lets you see THROUGH the object, and into some alternate dimension of some sort, and this obviously changes depending on the rotation of the camera.

Imagine an object you can look through that hides every other 3D object aside from the skybox.

NEW THREAD!!

>>57184318

>>57184296

But I was right, though. You do self-identify with some manner of communist leanings. Since you defy labels, you probably consider yourself some sort of enlightened reformist.

You're all the same.

>>57184299
Yes, exactly like a shitty greenscreen effect.

>>57184294
Effectively, instead of the model having a texture that's mapped to the model itself, it's just mapped to screenspace so it appears to operate like a "window" to the texture.

>>57184308
>>57184324
oh like a parallax effect for websites?

>>57184270
>>3
should know.

>>57184098

It would depend upon how much Facebook is paying more. Keep in mind that it is important to enjoy where you are working.

>>57184132

What Facebook product am I forced to use? Because I do not use any products by them, to my knowledge.

>>57184149

>Inheritance
The majority of millionaires and billionaires are self-made. Of those who are not, many may end up losing their fortunes if they are not careful with how they manage their money. Those who do not adapt a mentality of hard work will squander their inheritance.

>system abuse
You could still have a system where the 1% have more money than the 99% in a system where abuse does not take place. That said, there is no system where abuse does not take place, not even in non-capitalist systems.

>Why are you talking about labor as something one dimensional and not about specialty
It is about specialty. The problem is that numerous people don't try to learn the skills to specialize. People complain that the minimum wage is too low. No shit. You're not supposed to earn a living on that. You're supposed to learn a trade of some sort, or start a business if you can get the capital.

>There are more millionares in the world, then lets say a natural language processing specialists for AI or crisp biotech engineers who can actually benefit society
Because those natural language processing specialists are selling their skill, rather than an actual product. Rather than doing the labor of coming up with and marketing a product, they work for a company for a more stable wage, and let the company's other resources deal with whether or not the product they create succeeds or fails. The specialist laborer can expect the same amount of money if the product flops or not.

>>57184353
>>>/3/ god fukken dammit

Another stupid question: how do I extract the results, specifically the prediction and probability for each input from Tensorflow after running the model?

>>57184075
I have around 700 categories right now, one for each small feature and no crossed categories (I'm working on adding that now).

Don't have much training data since it needs to be processed by someone else, by hand, but a lot more is on the way.

Right now I only have 100 items.

>>57184144
That's what I figured. I'm largely using linear since there's a tutorial but I'll definitely look into changing it later on.

>>57184355
I understand your points, and trust me that is what I do if I want to continue in this capitalist economy.

But I am talking about ideals, and what is "fair". To which I response meritocracy is fair.

for example "the problem is that numerous people don't try to learn the skills to specialize", yet some that do exactly that are rich.

>>57184355
>The majority of millionaires and billionaires are self-made
Many of those that brag about it or are in public focus are self made.
And you most probably will hear of those that lost a lot.

You should take those two factors into consideration.

It's hard to lose your money in today's world. Rules are made and adapted by rich people to keep money to rich people, while keeping competition low.

>>57184400
>what is "fair".

Nothing will ever be fair, because people will never be truly equal.

>>57184428
Yeah, and meritocracy agrees with that even more brutally so than capitalism.

>>57184372
700 dimensions for 100 items... That's really off. I have never worked with such high dimensional but sparse data before - imagine you end up with 30 categories in the end but have exactly 3 examples for every category.

You should drop dimensions and check if the results get better. And get more data.

>>57184372
>>57184464
Yeah, if he even tried to make a vector space using his samples as a basis, he'd be missing 600 dimensions. And that's assuming his data is all linearly independent.

>>57184464
I asked about bucketization for related groups of features in each broad category, but my boss insists on not doing that.

>>57184270
If you want a static or animated image, you should use a cubemap and sample that as the texture.

If you want to have it realtime and dynamic and all that fun stuff, render the scene you want to a texture and sample that.

I'd recommend a cube map.

fuck, common lisp is goat
there's too much to take in though
reaching the macro chapter very soon

now to practice

>>57179646

Processes in an SSH session belong to your shell's process; typically:
sshd [your-username] tty[your-tty].

When you kill your session, all child processes are also exited.

To get around this, use terminal multiplexing, or daemons.

The most common multiplexers are screen and tmux.

You would then start screen by using the screen -S my-session-name, and running your process normally.

You can then press ctrl+a, d to detach from the screen, and the process will still be running, but as a child of screen, not your ssh session

To reattach, type screen -x my-session-name (-r also works)