approach also works in general with quadratic and even linear splines. First, the importing
of the respective sublibrary:
In [ 41 ]: import scipy.interpolate as spi
In [ 42 ]: x = np.linspace(- 2 * np.pi, 2 * np.pi, 25 )
We take again the original example function for illustration purposes:
In [ 43 ]: def f(x):
return np.sin(x) + 0.5 * x
The application itself, given an x-ordered set of data points, is as simple as the application
of polyfit and polyval. Here, the respective functions are splrep and splev. Table 9-2
lists the major parameters that the splrep function takes.
Table 9-2. Parameters of splrep function
Parameter Description
x
(Ordered) x coordinates (independent variable values)
y
( x-ordered) y coordinates (dependent variable values)
w
Weights to apply to the y coordinates
xb, xe
Interval to fit, if None [x[0], x[-1]]
k
Order of the spline fit ( 1 <= k <= 5)
s
Smoothing factor (the larger, the more smoothing)
full_output
If True additional output is returned
quiet
If True suppress messages
Table 9-3 lists the parameters that the splev function takes.
Table 9-3. Parameters of splev function
Parameter Description
x
(Ordered) x coordinates (independent variable values)
tck
Sequence of length 3 returned by splrep (knots, coefficients, degree)
der
Order of derivative (0 for function, 1 for first derivative)
