Scaling and normalization concepts in Data Analysis Python - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of Min-Max Scaling on a Simple Array

What is the output of the following code that applies Min-Max scaling to a numpy array?

Data Analysis Python

import numpy as np
from sklearn.preprocessing import MinMaxScaler

arr = np.array([[10], [20], [30], [40], [50]])
scaler = MinMaxScaler()
scaled_arr = scaler.fit_transform(arr)
print(scaled_arr.flatten())

A [0.2 0.4 0.6 0.8 1. ]

B [10. 20. 30. 40. 50.]

C [0. 0.25 0.5 0.75 1. ]

D [1. 0.75 0.5 0.25 0. ]

💡 Hint

Min-Max scaling transforms data to a range between 0 and 1 based on the minimum and maximum values.

Min-Max scaling subtracts the minimum value and divides by the range (max-min). Here, 10 maps to 0, 50 maps to 1, and values in between scale linearly.

❓ data_output

intermediate

Effect of Standardization on Data Mean and Std Dev

After applying standard scaling (z-score normalization) to this data, what are the mean and standard deviation of the transformed data?

Data Analysis Python

import numpy as np
from sklearn.preprocessing import StandardScaler

data = np.array([[5], [10], [15], [20], [25]])
scaler = StandardScaler()
scaled_data = scaler.fit_transform(data)
mean = scaled_data.mean()
std = scaled_data.std()
print(round(mean, 2), round(std, 2))

A 0.0 1.0

B 5.0 7.07

C 10.0 5.0

D 1.0 0.0

💡 Hint

StandardScaler centers data to mean 0 and scales to std dev 1.

StandardScaler subtracts the mean and divides by the standard deviation, resulting in data with mean 0 and std dev 1.

🔧 Debug

advanced

Identify the Error in Normalizing Data with sklearn

What error will this code raise when trying to normalize data using sklearn's Normalizer?

Data Analysis Python

from sklearn.preprocessing import Normalizer

X = [[1, 2], [3, 4], [5, 6]]
normalizer = Normalizer(norm='l3')
X_normalized = normalizer.transform(X)
print(X_normalized)

A No error, prints normalized array

B TypeError: 'list' object is not callable

C AttributeError: 'Normalizer' object has no attribute 'transform'

D ValueError: norm must be one of {'l1', 'l2', 'max'}

💡 Hint

Check the allowed norm parameters for Normalizer.

Normalizer only accepts 'l1', 'l2', or 'max' for the norm parameter. 'l3' is invalid and causes a ValueError.

🚀 Application

advanced

Choosing Scaling Method for Skewed Data

You have a dataset with a feature that is heavily skewed to the right (many small values, few very large values). Which scaling method is best to prepare this feature for a machine learning model?

A Log transformation followed by standardization

B Standardization (z-score) to center data with mean 0 and std dev 1

C Min-Max scaling to scale values between 0 and 1

D No scaling needed for skewed data

💡 Hint

Skewed data often benefits from transformations that reduce skewness before scaling.

Log transformation reduces skewness by compressing large values. Then standardization centers and scales the data for modeling.

🧠 Conceptual

expert

Understanding Differences Between Scaling and Normalization

Which statement correctly describes the difference between scaling and normalization in data preprocessing?

A Scaling adjusts data distribution shape; normalization adjusts data range

B Scaling changes data to a specific range; normalization changes data to have mean 0 and std dev 1

C Scaling and normalization are the same and can be used interchangeably

D Normalization changes data to a specific range; scaling changes data to have mean 0 and std dev 1

💡 Hint

Think about Min-Max scaling vs StandardScaler.

Scaling usually means adjusting data to a fixed range like 0 to 1 (Min-Max). Normalization often means standardizing data to mean 0 and std dev 1 (z-score).