Question

Can someone help me understand what we're being asked to do for Spring 2011, PS2 #1 "Implement the evaluate_poly function." Should our code start with asking for raw input for x and for a tuple of numbers for the poly, and then set up the function def evaluate_poly (poly, x)? I looked at the solution ps2_newton.py, but it just starts with setting up the function def evaluate_poly. When run, nothing happens because the solution code does not have any inputs. Did I misunderstand the assignment?

Answer 1

nope, they just want you to write the function - you don't have to 'ask' for inputs

Answer 2

Thanks. I'm completely new to programming and initially didn't get how to run the function in the python shell and the lecture screen shots are sometimes a little fuzzy. I figured out that after running it, in the python shell, we have to define poly, x, then type evaluate_poly(poly, x). Really basic, I know.

I was also wondering if there were a couple tiny errors in the PS2 #3 Newton solution set.
First, I tried running compute_root as is, without "return" statements in evaluate_poly and compute_deriv. The error is: line 11, in evaluate_poly for i in range(0,len(poly)): TypeError: object of type 'NoneType' has no len()

When I put in "return sumpoly" in evaluate_poly and "return derivTerm" in compute_deriv, this does not return an error. Is this because "return" somehow saves the answer for later use, but print does something else?

Second, in the function compute_deriv, should the range in "for" statement start at index 1, not 0 (as indicated in the solution set)? The result for compute_root is slightly different .80679 vs. .80678 and iterations of 7 vs. 140, but the code doesn't seem to be representing the math that should occur. poly=(-13.39, 0.0, 17.5, 3.0, 1.0) represents f(x)=x^4+ 3.0x^3 +17.5x^2 +0x - 13.39. Compute_deriv starting at index 0 results in a tuple of 5 terms: (-0.0, 0.0, 35.0, 9.0, 4.0). The Newton method entails a step to evaluate_poly on the derivative tuple, which will interpret the index as the exponent, ex: 4x^4, rather than 4x^3 because there are 5 terms. Is the best fix for this to start the range at index 1?

Third, I can't seem to get the 8 iterations from the example in the instructions. The closest I got was with using range starting at 1 indicated above, 7 iterations.

Thanks again.

Answer 3

a function will always return something even if you do not include a return statement. Without a return statement a function returns None.
return 'returns' an object to the thing that called the function

poly is a tuple of coefficients of a polynomial where the index of each item in the tuple is the power of that term in the function. the derivative of a five term poly should only have four terms - you lose the constant each derivative. so yep it should probably start at index one.

?with epsilon = .00001 you get 8 iterations?