Daily programming challenge

No fuck you.
I hate doing systems of equations on paper, I'm not gonna spend all afternoon writing your fucking homework.

>>59039175
How did you pass year 10?

>>59039192
idk lol

i never even took trig

>>59039206
Next /dpc/ will be on trig

>>59038960
wrong. 1, 3, and 5 should have the variable z, otherwise the question you're posing makes no sense (how can you output a three variable solution when you're only using two variables?).

Really, you've poorly defined the question. You should translate it like this:

1,2,3,4
translates to

1x+2y+3z+4=0

This whole eof thing is silly too and this is a poor way of implementing it into a problem to teach it to people. arguments should be sent to main or read from the input stream, no EOF is necessary (barring file reading), nor are commas, as that's up to your implementation choice.

>>59039360
I do like the idea of /dpc/ being a thing though, provided you're actually not just a bitch who can't do easy cs problems and is tryign to get g to do your homework

>>59039360
oh yes

>>59038891

4,-3,1,-10,EOF
2,1,3,0,EOF
-1,2,-5,17,EOF

translates to

4x-3y+z=-10
2x+y+3z=0
-x+2-5z=17

So the answer is: (x,y,z) = (1,4,-2)

Hey OP I'll play ball if you can explain how to solve systems of equations with matrices.

The example you posted introduces the number 17 out of nowhere, what the fuck.

>>59038891
Step 1: write system of equations in AX=B form
Step 2: look up and code how to inverse a matrix
Step 3: multiply inverse of A by B to get all X values

>>59039424
https://en.wikipedia.org/wiki/Gaussian_elimination

File: maxresdefault.jpg (62KB, 1920x1080px) Image search: [Google]

62KB, 1920x1080px

>>59039424
http://www.purplemath.com/modules/cramers.htm

>>59039503
>Gaussian elimination
I'd love to see an algorithm for that

>>59039502
Correction to my methods:
Step 1: write system of equations in AX=B form
Step 2: look up and code how to check determinants of matrices, if determinant not equal to 0: inverse matrix.
Step 3: look up and code how to inverse a matrix
Step 3: multiply inverse of A by B to get all X values

File: solver.png (4KB, 211x250px) Image search: [Google]

4KB, 211x250px

import Data.Matrix (Matrix, inverse, colVector, fromLists)
import Data.Vector (fromListN)

solve     m c = (* colVector c) <$> inverse m
inputMatrix i = fromLists       <$> mapM (const readLn) [1..i]
inputVector i = fromListN i     <$> readLn

main = do
  ps <- solve <$> inputMatrix 3 <*> inputVector 3
  either putStrLn print (ps :: Either String (Matrix Double))

not the same interface but you get the point

>>59039868
the Data.Matrix used is from the matrix package

>>59039868
why is your screenshot so small?

>>59039360
what's wrong with EOF?

>>59039507
This is actually really neat, thanks anon!

I'll post my solution when I come home.

>>59038891

#include <stdio.h>

typedef double vector3_t[3];

double det(vector3_t x, vector3_t y, vector3_t z) {
    return x[0]*y[1]*z[2] + x[1]*y[2]*z[0] + x[2]*y[0]*z[1] 
         - z[0]*y[1]*x[2] - z[1]*y[2]*x[0] - z[2]*y[0]*x[1];
}

int main() {
    vector3_t x, y, z, r;
    scanf("%lf,%lf,%lf,%lf, ", &x[0], &y[0], &z[0], &r[0]);
    scanf("%lf,%lf,%lf,%lf, ", &x[1], &y[1], &z[1], &r[1]);
    scanf("%lf,%lf,%lf,%lf, ", &x[2], &y[2], &z[2], &r[2]);
    double da = det(x, y, z);
    if (da == 0)
        printf("cannot solve\n");
    else
        printf("(%lf,%lf,%lf)\n", det(r, y, z)/da, 
                                  det(x, r, z)/da, 
                                  det(x, y, r)/da);
}

>>59040492
completely unnecessary in this use case. For example in C++ you can designate the delim character, so why the hell would you make the user type EOF (in whatever incarnation that takes in your operating system) when you #1 aren't using a file #2 are separating your inputs into three chunks.

It's a bad use case for EOF and concepts should be taught in use cases where they shine (provided said cases aren't extremely complex)

Do I win?

int det3(int mat[3][4])
{
    /* get 2x2 matrix determinants */
    unsigned i, j, r, x = 0, det = 0;
    for (r = 0; r < 3; r++) /* current row */
    {
        unsigned a[4], idx = 0;
        for (i = 0; i < 3; i++)
            for (j = 0; j < 3; j++)
                if (r != i && r != j)
                    a[idx++] = mat[i][j];
        int neg = (r == 1) ? -1 : 1;
        det += neg * (mat[r][x++] * ((a[0] * a[3]) - (a[1] * a[2])));
    }
    return det;
}

struct int3 systemofeq(int m[3][4])
{
    int d[3], det = det3(m);
    unsigned i, j;
    for (i = 0; i < 3; i++)
    {
        int z[3][4];
        unsigned k, l;
        for (k = 0; k < 3; k++) /* copy */
            for (l = 0; l < 4; l++)
                z[k][l] = m[k][l];
        for (j = 0; j < 3; j++) /* overwrite */
            z[j][i] = m[j][3];
        d[i] = det3(z);
    }
    struct int3 ans = { det / d[0], det / d[1], det / d[2] };
    return ans;
}

There are times I wish I didn't drop out of college. This is one of them.

>>59042766
No it's ok.
Knowing how to solve a system of equations is pretty useless in the real world.

>>59042899
kek

import numpy as np

inp = np.zeros((3, 4))
for i in range(3):
    inp[i, :] = eval(input())

mat = inp[:, :3]
vec = inp[:, 3]

np.linalg.solve(mat, vec)

File: Godfather.jpg (200KB, 1131x1600px) Image search: [Google]

200KB, 1131x1600px

[spoiler] bump [/spoiler]

Is this feels like cheating...

with Ada.Numerics.Generic_Real_Arrays;
with Ada.Text_IO; use Ada.Text_IO;

procedure System_of_Equations is
    package RA is new Ada.Numerics.Generic_Real_Arrays(Long_Float);
    use type RA.Real_Matrix;
    use type RA.Real_Vector;
    package Float_IO is new Ada.Text_IO.Float_IO(Long_Float);

    A : RA.Real_Matrix(1 .. 3, 1 .. 3) := (
        (7.0, 5.0, -3.0),
        (3.0, -5.0, 2.0),
        (5.0, 3.0, -7.0));

    B : RA.Real_Vector(1 .. 3) := (16.0, -8.0, 0.0);

    X_Linear_Algebra : RA.Real_Vector(1 .. 3) := RA.Inverse(A) * B;
    X_Idiot_No_Maths : RA.Real_Vector(1 .. 3) := RA.Solve(A, B);

begin

    -- uninteresting prints here

end;

>>59045541
>-- uninteresting prints here

>>59045555
But anon, it that part couldn't be less interesting.

>>59045695
isn't it supposed to accept runtime input?

File: a58.jpg (59KB, 590x589px) Image search: [Google]

59KB, 590x589px

>>59038891

A = [7 5 3; 3 -5 2; 5 3 -7]
b = [16; -8; 0]
A\b
done.

>>59038891
>Not using matlab

for j=1:3
    for i=1:4
        prompt='enter input';
        X(j,i)=input(prompt);
    end
end

X(1:3,1:3)\X(1:3,4)

http://pastebin.com/9PHuD5S0

Lack of real matrix libraries isn't going stop ME from wasting my time solving the general case.

Gaussian elimination.

gauss1 :: (Eq a, Fractional a) => [[a]] -> [[a]]
gauss1 m
  | length m <= 1 = m
  | otherwise = [m!!0] ++
    (map ([0]++) (gauss1 $ [[(m!!i!!j) - ((k!!i)*(m!!0!!j)) |
                   j <- cols, j > 0] | i <-rows, i > 0]))    
  where
    k = [(m!!i!!0)/(m!!0!!0) | i <- rows]
    cols = [0..((length $ m!!0)-1)]
    rows = [0..((length m) -1)]

gauss2 :: (Eq a, Fractional a) => [[a]] -> [a]
gauss2 m
  | length m == 1 = [(m!!0!!1)/(m!!0!!0)]
  | otherwise = [newsolution] ++ solutions
  where
    solutions = gauss2 [[m!!i!!j | j <- [0..(cols-1)], j > 0] | i <-[0..(rows-1)], i > 0]
    newsolution = ((m!!0!!(cols-1)) - sum [(m!!0!!j)*(solutions!!(j-1)) | j <- [1..(cols-2)]]) / (m!!0!!0)
    cols = length $ m!!0
    rows = length m

gauss = gauss2 . gauss1

>>59042899
You must not get those retarded FB posts that are basically linear algebra problems in picture form, or you've never been in a high paying field where applied math is the core of everything you do.

>>59048058
Nice

Problem is, that code is barely readable since I am not accustomed to functional programming

>>59050691
I bet it's terrible for those who are accustomed too, I just wanted it to work.