2520 is the smallest number that can be divided by each of the numbers from 1 to 10 without any remainder. Write a function called smallest_multiple that returns a uint64, the smallest positive number that is evenly divisible by all of the numbers fr

I keep getting an error everytime I run my code that says:

[out] = smallest_multiple(6) 
Error using gcd (line 23)
Inputs must be floats, namely single or double.

Error in smallest_multiple (line 4)
    out = (out * i) / (gcd(out,i));

Is 6 not a single?

Here is my code:

function [out] = smallest_multiple(n) 
out = uint64(1);
for i = 1 : n 
    out = (out * i) / (gcd(out,i));
end 
if out >= intmax('uint64')
    out = uint64(0); 
end 
end 

[Emma Sellers]

even if I do something like this:

function [out] = smallest_multiple(n) 
if n == uint64(1)
    out = uint64(1) ; 
elseif n == uint64(2)
    out = uint64(2) ; 
end 
end 

I still get an error:

Error using prod
PROD is only supported for floating point input. 

[Emma Sellers]

Which to my knowledge, i'm not using.

Is there some way in matlab to pre program variables to behave as certain data types?

eg, x(single) = 5 ;

[Stephen23]

"Is there some way in matlab to pre program variables to behave as certain data types? "

No, MATLAB is dynamically typed.

Also of interest:

https://www.mathworks.com/matlabcentral/answers/409283-2520-is-the-smallest-number-that-can-be-divided-by-each-of-the-numbers-from-1-to-10-without-any-rema

[Emma Sellers]

Thank you! I was usure of casting rules in matlab. I spent all of last semester on C programming and everything is running together.

[Emma Sellers]

Also, I tried running a code similar to the link you provided (Because he said it worked up through 16) and I STILL got the same error I've been getting and it didn't run at all.

Answer 1

[Jan]

Omit the explicit casting by uint64. Then Matlab uses the type double as default.

[Emma Sellers]

Ok, so now my code will run correctly if I manually run it. But I am still getting an error when I run the grader saying:

Error using prod
PROD is only supported for floating point input.

Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>sub_smallest_multiple
(line 696)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>run_submission
(line 510)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>grade
(line 431)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>grade_one
(line 328)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>hw8
(line 301)

[Emma Sellers]

Why would this code trigger a product of elements error? From line 4?

I can run the pieces individually with no problems.

[Jan]

If your function is wanted to reply an UINT64, cast the input for gcd:

for i = 1 : n 
    out = (out * i) / gcd(double(out), i);
end 

The message "PROD is only supported for floating point input" might mean, that there are more problems with the integer types in your code.

[Emma Sellers]

function [out] = smallest_multiple(n) 
out = uint64(1) ;
for i = 1 : n 
    out = uint64((out * i) / (gcd(double(out),i)));
end 
if out >= uint64(18446744073709551615)
    out = uint64(0); 
end 
end 

[Emma Sellers]

Can I do something similar to this? I'm still getting errors. I was unsure if I could cast a double inside something that is uint64

[Emma Sellers]

function [out] = smallest_multiple(n) 
out = uint64(1) ;
for i = 1 : n 
    y = (out * i) / (gcd(double(out),i));
    out = uint64(y);
end 
if out >= uint64(18446744073709551615)
    out = uint64(0); 
end 
end 

[Emma Sellers]

So, yes it needs to work with larger numbers.. But heres the thing:

when I manually run the code with 45 in it (the number its failing on) I get this error:

[out] = smallest_multiple(45)
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
out =
  uint64
   0

But, it still puts out 0, which is what the question ask for.. So im confused as to why it doesn't work. I even tried pre programming it to whenever n is greater than 41 it equals 0. Because thats when it exceeds the uint64 bounds.

Answer 2

[Rik]

There is an important difference between a single value (called a scalar in Matlab), and a value of data type single (which is an array that is stored in a specific format internally).

The gcd function requires the input to be a double or a single, so you should either keep it as a single/double or force it to be one, either with the cast function or by executing double(out).

[Emma Sellers]

But it wants the function to return a uint64.. So does that mean I can not use the greatest common denominator function?

[Rik]

You can also cast the result to uint64 after you've done the rest.

But this requirement does suggest to me that the intention is that you don't use it.

function [out] = smallest_multiple(n) 

%%%
%your code%
%%%

out=uint64(out);
end 

[Rik]

It seems you are making things more complicated than they need to be. Have a try at the code below:

clc
assert(isa(smallest_multiple(2),'uint64'),...
    'the assignment requires the ouput to be uint64')
for n=1:6
    fprintf('%d\n',smallest_multiple(n))
end

function [out] = smallest_multiple(in)
in = double(in); %input double type variable if it isn't already
out = 1;
for n = 1 : in %don't use i as a variable to avoid confusion with sqrt(-1)
    out = (out * n) / (gcd(out,n));
end
out = uint64(out); %convert to uint64 as required by the assignment
end

[Emma Sellers]

I still get the same prod error..

[Rik]

With my exact code? I must be missing something, because I don't see how that is possible. Can you put a breakpoint on the line that returns the error? Then, when you re-run the function, put

clc,whos

in the command window during the debugging and post the result here.

[Emma Sellers]

Well the code works to get expected answers - My code also worked to get expected answers when I ran it manually.

However, when I run your code (Copy and pasted from your post) It comes back with this error:

Problem 4 (smallest_multiple):
Error using prod
PROD is only supported for floating point input.

Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>sub_smallest_multiple
(line 696)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>run_submission
(line 510)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>grade
(line 431)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>grade_one
(line 328)


Error in
C:\Users\emm27219\Downloads\MATLAB-MOOC-HW\hw8\hw8.p>hw8
(line 301)

 
>

>

[Rik]

I think this is where I will need to give up. I cannot debug this if it is running from a .p file, as it is illegal to decompile/decrypt it (and sometimes not possible in the first place).

You can try using breakpoints in your function, but the p-file might not obbey them. I don't see any issue with the code (as we can run it without error outside of the p-file context), so I don't know what is going wrong here. I also don't know if there might be any issue with some function shadowing your implementation, nor do I know if there is a way to find out with a function like which.

The only solution I have is to ask your MOOC-instructor.

[Emma Sellers]

Turns out the error came from the fact that I was using an older version of matlab! So I had to switch to the online version. Now my code works until it tests '45' ?

[Rik]

Glad to hear you found a solution.

Do you need your code to work until a larger value? You might need to rework your algorithm. Another method of solving this question might be to check the prime factors and use ismember.

[Emma Sellers]

So, yes it needs to work with larger numbers.. But heres the thing:

when I manually run the code with 45 in it (the number its failing on) I get this error:

[out] = smallest_multiple(45)
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
out =
  uint64
   0

But, it still puts out 0, which is what the question ask for.. So im confused as to why it doesn't work. I even tried pre programming it to whenever n is greater than 41 it equals 0. Because thats when it exceeds the uint64 bounds.

Answer 3

[Emma Sellers]

So, yes it needs to work with larger numbers.. But heres the thing:

when I manually run the code with 45 in it (the number its failing on) I get this error:

[out] = smallest_multiple(45)
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 
Warning: Inputs contain values larger than the largest consecutive flint.
         Result may be inaccurate. 
> In gcd (line 64)
  In smallest_multiple (line 6) 

out =

  uint64

   0

But, it still puts out 0, which is what the question ask for.. So im confused as to why it doesn't work. I even tried pre programming it to whenever n is greater than 41 it equals 0. Because thats when it exceeds the uint64 bounds.

[Steven Lord]

The smallest number that can be divided by each of the numbers from 1 to 45 is larger than the largest value that can be stored as an int64. If you switch to a uint64 you can get up to 1 to 46, but 1 to 47 is too large even for that.

If you want to go further than that, you're probably going to want to switch over to using Symbolic Math Toolbox.

[Emma Sellers]

I'm currenly using "uint64" but it still says that the number exceeds bounds?

[Emma Sellers]

What is the symbolic math toolbox?

[Emma Sellers]

My code is at the very top of this feed. When I run the code like this:

[out] = smallest_multiple(45)

out =

  uint64

   409927646082434480

Thats what I get, but the grader STILL says it wrong cries*** ive been trying to figure this out for days.

[Steven Lord]

That answer is not correct. There are several ways to see this. The first is to try the division.

x = uint64(409927646082434480)
d = x/3;
d2 = 3*d;
check = [x; d2]

If x were divisible by 3, the two elements in check would be the same. They aren't.

The sum of the digits in x is 80, which is another sign the answer is incorrect. The divisibility rule for 3 is that the sum of the digits must itself be a multiple of 3.

A third check is that the result must be divisible by both 25 (=5^2) and 4 = (2^2) so it must be a multiple of 100 (=2^2 * 5^2).

At this point, I think you've done your due diligence and should probably talk to your professor or teaching assistant. Do they expect you to do the calculations symbolically and check afterward if it's too large to fit in a 64-bit integer? Do they expect you to perform the calculations iteratively and stop as soon as the next one would saturate? It's not quite clear from the snippet of the assignment in the title of this post.