This notebook demonstrates step-by-step data wrangling techniques using the Titanic dataset.
The main tasks include exploring, cleaning, transforming, grouping, and merging data.

🔗 Useful Links

• 👉 Open in Google Colab
  OR
• Open in Github
• 👉 GitHub Titanic Dataset

1. Getting Information about the Data

• Load dataset (read_csv).
• Preview data using head().
• Get dataset dimensions with shape.
• Generate descriptive statistics with describe().

2. Slicing a DataFrame

• Select rows by position (iloc).
• Set an index (set_index).
• Select rows by label (loc).

3. Selecting Rows Based on Conditionals

• Filter rows using conditions (df[df['Sex']=='female']).
• Combine multiple conditions with & (e.g., female passengers aged ≥65).

4. Sorting Values

• Sort rows using sort_values().
  Example: sorting passengers by age.

5. Replacing Values

• Replace specific values (replace({'male':'M'})).
• Use regex replacement (replace(r"male","third",regex=True)).

6. Renaming Columns

• Rename columns using rename().
  Example: renaming Sex → Gender.

7. Finding Maximum, Minimum, Sum, Average, Count

• Use aggregation functions like max(), min(), mean(), sum(), and count() on a column.

8. Finding Unique Values

• Get unique values in a column (unique()).
• Count occurrences (value_counts()).

9. Handling Missing Data

• Identify missing values (isnull() / isna()).
• Replace values with NaN (replace + numpy.nan).
• Fill missing values with statistics (e.g., mean age using fillna()).

10. Deleting Columns

• Drop columns with drop(axis=1).

11. Deleting Rows

• Remove rows conditionally (df[df['Sex']!='male']).
• Drop specific values (e.g., removing age = 65).

12. Dropping Duplicate Rows

• Remove duplicates using drop_duplicates().
• Count duplicates with duplicated().sum().

13. Group Rows by Values

• Group rows with groupby().
• Apply aggregations (e.g., count per Sex).

14. Grouping Rows by Time

• Create a time index using pd.date_range().
• Resample by time period (W = weekly, M = monthly).
• Perform aggregations (sum, mean) on grouped data.

15. Aggregating Operations and Statistics

• Use aggregate() with multiple operations.
  Example: Age min & max, Fare mean.

16. Looping Over a Column

• Iterate through column values with a loop.
• Example: converting first two names to uppercase.

17. Applying a Function Over a Column

• Apply a custom function with apply().
  Example: convert names to uppercase.

18. Applying a Function to Groups

• Apply functions to grouped data.
  Example: groupby('Sex').apply(lambda x: x.count()).

19. Concatenating DataFrames

• Combine DataFrames vertically with pd.concat().

20. Merging DataFrames

• Merge DataFrames on a common key using pd.merge().
  Example: merging employees with sales data.

✅ Final Notes

This notebook demonstrates key pandas techniques for data wrangling:

• Data exploration & cleaning.
• Slicing, filtering, sorting.
• Handling missing data & duplicates.
• Aggregating & grouping.
• Combining multiple DataFrames (concat & merge).

Together, these operations form the foundation of data preprocessing and analysis in Python.

Jai Hanuman