New with Python. Help with a code

[jasonkeller1970]

Hi,
I just registered in this site because I would like to know if it is possible to solve this problem with a simple Python code.
I'm not into programming and my knowledge is very low; I don't even know if it is possible to have a simple code for this problem.
I'm a mathematician and I have to compute a certain function (that has a certain number of variables) with the variables running in a set of values. Is it possible to have a code that print on the screen the values of the function corresponding to the set of variables?
I try to be more clear making an easy example.
Let suppose that I declare the function z=2 * x**4 - 5 * y**3 and I would like that the code prints on the screen all the values of z when, for instance, x and y goes from 1 to 5 (entire values). I expect that the screen shows me the corresponding 25 values of z.
As you can see I don't have any knowledge; I could suppose that this is a really simple problem even if I don't know how to begin.
Can somebody help me?
Thank you
Jason

[deanhystad]

I can think of a dozen ways to do this. Picking the right one depends on how you use it.

This is the simplest

for x in range(1, 6):
    for y in range(1, 6):
        print("x =", x, "y =", y, "z =", 2 * x**4 - 5 * y**3)

Does that do what you want?

[jasonkeller1970]

I can think of a dozen ways to do this. Picking the right one depends on how you use it.

This is the simplest

for x in range(1, 6):
    for y in range(1, 6):
        print("x =", x, "y =", y, "z =", 2 * x**4 - 5 * y**3)

Does that do what you want?

I test it and it is perfect! Thank you!
I suppose that if the variables are for instance 10 I cannot use a wide range for each of them.......I mean for the computation time.
In any case, if I would like to get only the negative values of the function there is a command that print only the negative values excluding the positive ones?

[deanhystad]

The values for x and y can be anything you want. I used range(1, 6) because you specified a "1 to 5" and "25 corresponding values". You could use numpy arrays if you wanted finer resolution.

import numpy as np
x = np.linspace(1, 5, 501)  # list of numbers 1.0, 1.1, 1.2 ... 4.8, 4.9, 5.0

You could solve for thousands of points per axis, but as the number of results gets in the millions or larger you will notice some lag. But if you are looking millions of points, you can no longer use print and have to start looking at plotting the results using something like matplotlib.

If you want to do a lot of numerical analysis you should learn about numpy, pandas and matplotlib

https://numpy.org/doc/stable/user/absolu...nners.html
https://pandas.pydata.org/pandas-docs/st...10min.html
https://www.geeksforgeeks.org/python-int...atplotlib/

[jasonkeller1970]

The values for x and y can be anything you want. I used range(1, 6) because you specified a "1 to 5" and "25 corresponding values". You could use numpy arrays if you wanted finer resolution.

import numpy as np
x = np.linspace(1, 5, 501)  # list of numbers 1.0, 1.1, 1.2 ... 4.8, 4.9, 5.0

You could solve for thousands of points per axis, but as the number of results gets in the millions or larger you will notice some lag. But if you are looking millions of points, you can no longer use print and have to start looking at plotting the results using something like matplotlib.

If you want to do a lot of numerical analysis you should learn about numpy, pandas and matplotlib

https://numpy.org/doc/stable/user/absolu...nners.html
https://pandas.pydata.org/pandas-docs/st...10min.html
https://www.geeksforgeeks.org/python-int...atplotlib/

Thank you for your help. The kind of values are not important. I try to explain. My field is not numerical analysis....I'm working on a theoretic problem.
I have to prove that an expression that depends of a certain numbers of variables is nonnegative. The variables represent the values of the derivative (of first, second and third order) of an unknown function. I'm looking for some assumptions on the function that assure that the very long expression is always nonnegative for every choice of the parameters. The expression comes from the determinant of a square matrix (I have a question about this matter.....I ask at the end of the post) that become bigger and bigger with the number of variables of the function. Obviously at the end I have to prove that my expression is nonnegative for EACH choice of parameters and obviously it is impossible to test the expression in all the possible values. But when I find that with some assumption on the function I get always nonnegative value with the help of python script, I could try to work on the expression to prove that is nonegative just working on its terms (for instance showing that it is sum of squares). On the contrary, if the python script show me negative values obviously it is useless working on the expression. That's why the script you gave me in enough! Thank you!

Another question anyway. As I have written, my expression is the result of a determinant of a square matrix so it could be useful for me define a matrix. I have only one question. I have found with google this example to calculate the determinant of a matrix

import numpy
m=numpy.array([[2,-1,0],[1,1,1],[0,1,-1]])
k=numpy.linalg.det(m)
print(k)

anyway when I use this code I don't get the correct result (that is -5) but I get -4.9999999998. There is a logic reason for this?

buran write Apr-18-2023, 12:29 PM:
Please, use proper tags when post code, traceback, output, etc. This time I have added tags for you.
See BBcode help for more info.

[buran]

Why are floating-point calculations so inaccurate?

Is floating point math broken?

[Gribouillis]

I have found with google this example to calculate the determinant of a matrix

numpy.linalg.det() leverages the well known LAPACK library (compiled Fortran code) which uses LU decomposition for this task.

Efficient exact computation of an integer determinant is not trivial. See a pure Python implementation of the Bareiss algorithm here

[jasonkeller1970]

I noticed something strange when I use the code but probably I make something wrong. I have different question about:

When I start this code

for a in range(1, 4):
    for b in range(1, 6):
        for c in range(5, 8):
            for d in range(1, 5):
                for e in range(5, 8):
                    for f in range(5, 8):
                        for g in range(1, 10):
                            for h in range(1, 3):
                                for l in range(1, 4):
                                    result = (c**2+a*f+d**2+b*g)*(d**2+a*h+e**2+b*l)-(c*d+a*g+d*e+b*h)**2
                   print("a =", a, "b =", b, "c =", c, "d =", d, "e =", e, "f =", f, "g =", g, "h =", h, "l =", l, "z =", result)

i don't get all the result probably because the terminal shows only a limited number of outputs. Is this true? There is a way to get all the results in a text file?

Anyway, the more serious problem for me it this one: as I wrote in a precedent post I'm interested only in the negative results. If I scroll up the list of the outputs I see the value -233. But if I use the same code as before just adding before the last line the instruction

if result < 0:

I get some negative outputs but in the list -233 is missing. Why? Am I doing something wrong?

[deanhystad]

Inbdentation for the print is wrong. It should be deeper than any other line.
That is why you are missing values in the output. As is, the program only prints f=8, g=10, h=3, l=4 values. When you added the if statement you likely messed up the indentation in a differenr way that lead to more or fewer results being printed.

[Gribouillis]

There is a way to get all the results in a text file?

This code will write all the results in a comma-separated values file named 'output.csv' that you can open with a spreadsheet application.

import csv
import itertools as itt

def generate_results():
    for a, b, c, d, e, f, g, h, l in itt.product(
        range(1, 4), range(1, 6), range(5, 8), range(1, 5), range(5, 8),
        range(5, 8), range(1, 10), range(1, 3), range(1, 4)):
        result = ((c**2+a*f+d**2+b*g)*(d**2+a*h+e**2+b*l)
                -(c*d+a*g+d*e+b*h)**2)
        yield [a, b, c, d, e, f, g, h, l, result]


with open('output.csv', 'w') as csvfile:
    fieldnames = 'a b c d e f g h l r'.split()
    writer = csv.writer(csvfile)
    writer.writerow(fieldnames)
    for item in generate_results():
        writer.writerow(item)

The value -233 is in the output file.