BSpline process from R to Python

Question

I have a some code that performs smoothing process using BSpline and fdPar functions. Below is the process:

Start to process

1. norder = 6, which is 1 more than degree of Bspline.

2. Setup the spline base Object - basisobj.

rangval <- c(1, length(data))

breaks = seq(1,length(data),length.out=length(data)/60)

nbasis = length(breaks) - 2 + norder

basisobj = create.bspline.basis(rangval,nbasis,norder=norder,breaks=breaks)

3. Define a Functional Parameter Object - fdParobj

fdParobj = fdPar(basisobj)

4. Define a fdobj

fdobj = smooth.basis(timepoints,data,fdParobj)

5. predict

I am trying to convert the code / process above into Python.

I have been able to perform the BSpline to some extent. See below:

from scipy.interpolate import BSpline

import matplotlib.pyplot as plt

import pandas as pd

import numpy as np

import math

breaks = np.linspace(1, (len(pressure_data)), math.ceil(len(pressure_data)/60))

k = math.ceil(len(pressure_data)) - 2

degree = 5

order = degree + 1

n = order + k

t = np.zeros(math.ceil(len(pressure_data)/60) + (2*order)) # create an array for knots

t[:order] = 0

t[-order:] = len(pressure_data)

t[order:-order] = (breaks )

xx = np.arange(len(pressure_data))

for i in range(0,n):

c = np.zeros(n)

c[i] = 1

spl = BSpline(t,c,degree)

plt.plot(xx, spl(xx))

plt.show()

However, I am trying to figure out how to convert the remaining (Step 3 and 4) into Python. Any help will be much appreciated.