How to find factorial(c) of an array of numbers?

Question

surabhi sachdeva 2020-5-26

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/534648-how-to-find-factorial-c-of-an-array-of-numbers

评论： surabhi sachdeva 2020-5-27

Respected all

I have successfully got the values for j and v. I have got stuck in solving t due to the problem in finding factorial(c). My question is how to code for factorial term in the expression? Is it possible or I am on the wrong path? As without solving factorial, t cannot be determined. What should I do?

Code attached:

A = linspace(1,10)

C = linspace(1, 11)

[k, c] = ndgrid(A, C)

j=gammainc(c,(0.85));

v=gamma(c.^(-1));

t=0.85.*(1-(((0.85).^5.*(1-(((0.85))./(1-(((0.85)).^6)))))).*1.*(1-((0.85).^5.*(factorial(c)).^-1.*(c.^(5-c).^-1)./(exp(0.85).*k.*v+(0.85).^c.*(1-(0.85./c).^(5-c+1)))/factorial(c).*(1-(0.85./c)))));

The following error is being encountered

Error using factorial

N must be an array of real non-negative integers.

How can this error be removed?

Please help

Regards

Surabhi

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Adam Danz 2020-5-26

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/534648-how-to-find-factorial-c-of-an-array-of-numbers#answer_439643

编辑：Adam Danz 2020-5-26

N must be an array of real non-negative integers.

To troubleshoot that error message, let's go through the list of possibilities.

% Is c an array or real numbers?
all(isreal(c(:)))  % = 1, true
% Is c all non-negative?
all(c(:)>=0)  % = 1, true
% Is c all integers?
all(isinteger(c(:)))  % = 0!  FALSE

There's the problem. C contains non-integers. It makes sense that this would cause an error, right? Factorials are functions of intergers.

factorial(3) = 24

4 * 3 * 2 * 1 = 24

How could that be done with non-integers?

9 个评论
显示 7更早的评论隐藏 7更早的评论

Adam Danz 2020-5-27

编辑：Adam Danz 2020-5-27

There are many moving parts in that very long equation. The NaNs are not being caused by the factorial() output.

There are some key points to understand about this new problem.

Understand these points before solving the new problem.

1) There's probably several ways to change the equation so that non-NaN values are returned but that doesn't mean the changes that were made are correct. For example, (1+inf)/(1-inf) returns a NaN. I could replace either of the 2 inf values with a non-inf number or I could replace both of them and the result will not be nan but I don't know what the equation is supposed to be.

2) The problem could be that the t equation is fine but the inputs are the wrong size or shape. For example, if the inputs are scalar instead of arrays, the currect equation works!

k = 1;
c = 2; 
j = 0.896; 
v = 1.7725; 
t=0.85.*(1-(((0.85).^5.*(1-(((0.85))./(1-(((0.85)).^6)))))).*1.*(1-((0.85).^5.*(factorial(c)).^-1.*(c.^(5-c).^-1)./(exp(0.85).*k.*v+(0.85).^c.*(1-(0.85./c).^(5-c+1)))/factorial(c).*(1-(0.85./c)))))
% RESULT:
% t =  0.98732

3) The problem could be that the t equation was coded incorrectly. I have a feeling this is the problem. Specifically, I think some of the array operations should be element-wise operations instead of matrix/vector operations. To understand this difference, read this page of the documentation. <-- don't skip this part, it's the most important part of the entire answer.

For example, adding this period results in non-NaN output when inputs are arrays.

t=0.85.*(1-(((0.85).^5.*(1-(((0.85))./(1-(((0.85)).^6)))))).*1.*(1-((0.85).^5.*(factorial(c)).^-1.*(c.^(5-c).^-1)./(exp(0.85).*k.*v+(0.85).^c.*(1-(0.85./c).^(5-c+1)))./factorial(c).*(1-(0.85./c)))))
% LOOK HERE ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> ---> --> ^ 

But that doesn't mean the this is the correct solution. That's up to you to think about.