how to do a multi dimensional function fitting using python

Question

I am doing some least square fitting things. and it's two dimensional which means (x1i,x2i)-->(yi).So far i checked a lot online documents which are designed for 1 dimensional (xi)->(yi).

1 So anyone knows how to solve multi-dimensional least fitting using any standard packages? such as leastsq?

2 And the other question is how to return multiple functions!!! The reason I want to do this is just dealing with the yi-f(xi) part. If i have limited dataset, i can do something shown below. But How could i do it with a lot data? I am not able to write every individual function by hand...

def F(x):
return (weight(x[0],x[1],x[2],weeknumbers[0])-13,
        weight(x[0],x[1],x[2],weeknumbers[1])-21,
        weight(x[0],x[1],x[2],weeknumbers[2])-28,
        weight(x[0],x[1],x[2],weeknumbers[3])-36,
        weight(x[0],x[1],x[2],weeknumbers[4])-36,
        weight(x[0],x[1],x[2],weeknumbers[5])-39,
        weight(x[0],x[1],x[2],weeknumbers[6])-39,
        weight(x[0],x[1],x[2],weeknumbers[7])-41,
        weight(x[0],x[1],x[2],weeknumbers[8])-42)

You might find numpy helpful. A lot of linear algebra functions already exist there. — John Powell, Apr 28 '14 at 06:51

score 0 · Answer 1 · edited May 23 '17 at 12:28

0

For input you can declare multiple arguments "def F(x,y,z):" for return you can use tuples (How do you return multiple values in Python?) or just return a table: "return [a,b,c]"

edited May 23 '17 at 12:28

Community

1
1

answered Apr 28 '14 at 06:47

hegelsturm

71
1
6

how to do a multi dimensional function fitting using python

1 Answers1