Fitting custom models in SciPy - Mini Project: Build & Apply

Fitting Custom Models with SciPy

📖 Scenario: You work as a data analyst for a small company that collects data on how the temperature affects the sales of ice cream. You want to find a simple mathematical model that fits the sales data so you can predict future sales based on temperature.

🎯 Goal: Build a custom mathematical model and fit it to the given temperature and sales data using SciPy's curve fitting tools. Then, display the fitted model's predicted sales for the given temperatures.

📋 What You'll Learn

Create a dataset of temperatures and corresponding ice cream sales.

Define a custom model function that relates temperature to sales.

Use SciPy's curve_fit function to find the best parameters for the model.

Print the predicted sales using the fitted model.

💡 Why This Matters

🌍 Real World

Fitting custom models helps businesses understand relationships in their data, like how temperature affects sales, so they can make better decisions.

💼 Career

Data scientists and analysts often fit models to data to predict outcomes and find trends, which is essential in many industries.

1

Create the temperature and sales data

Create two lists called temperatures and sales with these exact values: temperatures = [15, 18, 21, 24, 27, 30] and sales = [100, 150, 200, 250, 300, 350].

SciPy

# Create the temperature and sales data lists
# Your code here

Need a hint?

Use square brackets to create lists and separate values with commas.

2

Define a custom linear model function

Define a function called linear_model that takes three parameters: x, a, and b. The function should return the value of a * x + b.

SciPy

temperatures = [15, 18, 21, 24, 27, 30]
sales = [100, 150, 200, 250, 300, 350]

# Define the linear_model function
# Your code here

Need a hint?

Define a function using def and return the expression a * x + b.

3

Fit the linear model to the data

Import curve_fit from scipy.optimize. Use curve_fit with the linear_model function, temperatures, and sales to find the best parameters. Store the parameters in a variable called params.

SciPy

temperatures = [15, 18, 21, 24, 27, 30]
sales = [100, 150, 200, 250, 300, 350]

def linear_model(x, a, b):
    return a * x + b

# Import curve_fit and fit the model
# Your code here

Need a hint?

Use from scipy.optimize import curve_fit and call curve_fit(linear_model, temperatures, sales).

4

Print the predicted sales using the fitted model

Use the fitted parameters in params to calculate predicted sales for each temperature in temperatures by calling linear_model. Store the results in a list called predicted_sales. Then print predicted_sales.

SciPy

from scipy.optimize import curve_fit

temperatures = [15, 18, 21, 24, 27, 30]
sales = [100, 150, 200, 250, 300, 350]

def linear_model(x, a, b):
    return a * x + b

params, _ = curve_fit(linear_model, temperatures, sales)

# Calculate predicted sales and print them
# Your code here

Need a hint?

Use a list comprehension to apply linear_model to each temperature with the fitted parameters, then print the list.