error loops, Gaussian elimination julia - system

Question

the code below solves a linear system using Gauss method. when I run again, the error occurs:
LoadError: InexactError()
while loading In[176], in expression starting on line 4
in setindex! at array.jl:313
[inlined code] from In[176]:13
in anonymous at no file:12
Running more than 3 times the system is solved. what's up?
A = [4.1 -5 6.8;7.8 -8 -9;-17 4 1];
b = [1,2,3];
x = zeros(3);
m = 0;
al,ac = size(A)
for k= 1:(al-1)
#println("valor de k ",k)
for i = (k+1):al
#println("valor de i ",i)
m = A[i,k]/(A[k,k])
A[i,k] = 0
for j=(k+1):al
#println("valor de j",j)
A[i,j] = A[i,j] - m*A[k,j]
b[i]= b[i] - m*b[k]
end
end
end
x[al] = b[al]/(A[al,al])
for k = (al-1):-1:1
begin
s = 0;
for j = (k+1):al
s = s+A[k,j]*x[j]
end
x[k]=(b[k]-s)/A[k,k]
end
end
println(x)

Answer 1

The error is occurring mainly due to type mismatch between the variable and the value, consider the simple example,
julia> convert(Int64, 0.1)
ERROR: InexactError()
The InexactError() is thrown because it not possible to represent 0.1 as an integer.
julia>convert(Int64, 1.0)
1
As you can see the value 1.0 does not have problems being converted to Int64. So in your case since the type for array b is Array{Int64,1} and it sure cant hold floating point variables in due process and hence the error.
Considering #jeff 's inputs, it might be better to pakckthe code in a function,
julia>function f(A::Array{Float64,2}, b::Vector{Float64})
x = zeros(3);
m = 0.0;
al,ac = size(A)
for k= 1:(al-1)
#println("valor de k ",k)
for i = (k+1):al
#println("valor de i ",i)
m = A[i,k]/(A[k,k])
A[i,k] = 0
for j=(k+1):al
#println("valor de j",j)
A[i,j] = A[i,j] - m*A[k,j]
b[i]= b[i] - m*b[k]
end
end
end
x[al] = b[al]/(A[al,al])
for k = (al-1):-1:1
begin
s = 0;
for j = (k+1):al
s = s+A[k,j]*x[j]
end
x[k]=(b[k]-s)/A[k,k]
end
end
return x, A, b
end
f (generic function with 1 method)
julia> A = [4.1 -5 6.8;7.8 -8 -9;-17 4 1];
julia> b = [1,2,3.0];
This should result in the following transformation of matrix A into a upper triangular form,
julia> A
3×3 Array{Float64,2}:
4.1 -5.0 6.8
0.0 1.5122 -21.9366
0.0 0.0 -213.523