UnivariateSpline in SciPy - Step-by-Step Execution

Concept Flow - UnivariateSpline

Input data points x, y

↓

Create UnivariateSpline object

↓

Fit spline curve to data

↓

Evaluate spline at new points

↓

Get smooth curve output

We start with data points, create a spline object that fits a smooth curve, then evaluate it to get smooth values.

Execution Sample

SciPy

import numpy as np
from scipy.interpolate import UnivariateSpline

x = np.linspace(0, 10, 10)
y = np.sin(x)
spline = UnivariateSpline(x, y, s=1)
y_smooth = spline(np.linspace(0, 10, 100))

This code fits a smooth spline to sine data and evaluates it at 100 points for a smooth curve.

Execution Table

Step	Action	Input/Condition	Result/Output
1	Generate x points	np.linspace(0,10,10)	[0.0, 1.11, ..., 10.0] (10 points)
2	Calculate y = sin(x)	sin(x)	[0.0, 0.9, ..., -0.54] (10 points)
3	Create UnivariateSpline	x, y, s=1	Spline object created, smoothing applied
4	Evaluate spline at 100 points	np.linspace(0,10,100)	Array of 100 smooth y values
5	Output smooth curve	y_smooth array	Smooth sine curve values for plotting

💡 All steps complete, spline fitted and evaluated successfully

Variable Tracker

Variable	Start	After Step 1	After Step 2	After Step 3	After Step 4	Final
x	None	[0.0, 1.11, ..., 10.0]	[0.0, 1.11, ..., 10.0]	[0.0, 1.11, ..., 10.0]	[0.0, 1.11, ..., 10.0]	[0.0, 1.11, ..., 10.0]
y	None	None	[0.0, 0.9, ..., -0.54]	[0.0, 0.9, ..., -0.54]	[0.0, 0.9, ..., -0.54]	[0.0, 0.9, ..., -0.54]
spline	None	None	None	Spline object	Spline object	Spline object
y_smooth	None	None	None	None	[smooth values]	[smooth values]

Key Moments - 3 Insights

Why do we use the parameter s=1 when creating UnivariateSpline?

The s parameter controls smoothing. Setting s=1 allows some smoothing to avoid fitting noise exactly. See execution_table step 3 where spline is created with s=1.

What happens if we evaluate the spline at points outside the original x range?

Evaluating outside original x can extrapolate but may be unreliable. Our example evaluates within range (step 4).

Why do we use more points (100) to evaluate the spline than original data points (10)?

Using more points creates a smooth curve for plotting, showing the spline's shape clearly (step 4 and 5).

Visual Quiz - 3 Questions

Look at the execution_table, what is the output of step 2?

A Spline object created

B Array of 10 sine values corresponding to x

C Array of 100 smooth y values

D Input x points

💡 Hint

Check the 'Result/Output' column for step 2 in execution_table

At which step is the smoothing parameter s=1 applied?

A Step 4

B Step 1

C Step 3

D Step 5

💡 Hint

Look at the 'Action' and 'Input/Condition' columns in execution_table step 3

If we change the number of evaluation points from 100 to 50, what changes in variable_tracker?

A y_smooth array length changes to 50

B x array length changes to 50

C Spline object changes

D Original y values change

💡 Hint

Check the 'y_smooth' variable in variable_tracker after step 4

Concept Snapshot

UnivariateSpline fits a smooth curve through data points.
Syntax: UnivariateSpline(x, y, s=smoothing_factor)
's' controls smoothness; s=0 fits all points exactly.
Evaluate spline by calling it with new x values.
Useful for smoothing noisy data and interpolation.

Full Transcript

We start with input data points x and y. Then we create a UnivariateSpline object with smoothing parameter s=1 to fit a smooth curve. Next, we evaluate this spline at many points to get a smooth output curve. Variables x and y hold original data, spline holds the fitted model, and y_smooth holds the smooth curve values. Key points include understanding smoothing parameter s, evaluating spline within data range, and using more points for smooth plotting.