Lesson 16: Final Review & Comprehensive Practice

Final Review – Data Handling & Visualization

Congratulations! You’ve completed Module 2.

Recap of key concepts:

Pandas Series & DataFrames – 1D & 2D data
Loading data (CSV, Excel, JSON)
Exploring: head, info, describe
Missing values: isnull, dropna, fillna
Data cleaning: types, duplicates, renaming
Filtering & sorting
Grouping & aggregation
Merging & joining
Matplotlib basics (plots, customization)
Seaborn: beautiful statistical plots
Box, heatmap, pair plots
Mini-project: Titanic exploration

You can now load, clean, explore, and visualize real datasets — ready for modeling!

Exercise 1

Exercise 2

Match the Pandas/Seaborn tool to its use (drag to first empty zone):

df.describe() → Drop here
sns.heatmap() → Drop here
df.groupby().mean() → Drop here
df.dropna() → Drop here

Click / tap any item to place it in the first empty zone:

Statistical summary of numeric columns

Correlation matrix visualization

Average per category

Remove rows with missing values

Exercise 3

Which are true about data handling in ML? (Select all)

80% of ML time is data preparation Visualization helps spot outliers & patterns Pandas is built on NumPy Seaborn is better for quick statistical plots You can skip cleaning if model is powerful

← Previous Lesson (15) Next Lesson →