Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #42120 > unrolled thread

Differentiation in Python

Back to article view | Back to comp.lang.python

Contents

  Differentiation in Python zingbagbamark@gmail.com - 2013-03-28 05:17 -0700
    Re: Differentiation in Python Steven D'Aprano <steve+comp.lang.python@pearwood.info> - 2013-03-28 12:35 +0000
    Re: Differentiation in Python David Robinow <drobinow@gmail.com> - 2013-03-28 14:16 -0400

#42120 — Differentiation in Python

How do I differentiate(first order and 2nd order) the following equations in python. I want to differentiate C wrt Q.

C = (Q**3)-15*(Q**2)+ 93*Q + 100

[toc] | [next] | [standalone]

#42121

On Thu, 28 Mar 2013 05:17:36 -0700, zingbagbamark wrote:

> How do I differentiate(first order and 2nd order) the following
> equations in python. I want to differentiate C wrt Q.
> 
> C = (Q**3)-15*(Q**2)+ 93*Q + 100


For such a simple case, you don't do it with Python, you do it with 
maths, and get an exact result.

dC/dQ = 3*Q**2 - 30*Q + 93


d²C/dQ² = 6*Q - 30


The rule for differentiating polynomials of the form

y = k*x**n

is:

dy/dx = (k*n)*x**(n-1)

Consult a good calculus text book or on-line site for further details.


-- 
Steven

[toc] | [prev] | [next] | [standalone]

#42178

On Thu, Mar 28, 2013 at 8:17 AM,  <zingbagbamark@gmail.com> wrote:
> How do I differentiate(first order and 2nd order) the following equations in python. I want to differentiate C wrt Q.
>
> C = (Q**3)-15*(Q**2)+ 93*Q + 100

"""
 Years ago, when I actually worked for a living, I would have
done something like this:
"""
coeffs = [1, -15, 93, 100]
num_coeffs = len(coeffs)-1
deriv = [coeffs[i]*(num_coeffs-i) for i in range(num_coeffs)]
print(deriv)

"""
The above is somewhat obscure and requires one to add
some documentation.  Who wants to do that?

Below is a version using numpy. You get the numpy docs
for free.
"""
import numpy as np
p = np.poly1d(coeffs)
deriv_np = np.polyder(p)
print(deriv_np)
# or
print(list(deriv_np))

[toc] | [prev] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web