Pandas course

Lesson 1 • Core Python Data Structures

Review lists, dictionaries, and tuples as the building blocks Pandas extends. Connects native Python containers to Pandas Series and DataFrame concepts.

Lesson 2 • NumPy Arrays as Pandas Backbone

Introduce NumPy arrays and vectorized operations that power Pandas internally. Understanding arrays prevents confusion when Pandas returns array-based results.

Lesson 3 • Setting Up the Environment

Install Python, Pandas, and Jupyter Notebook for an interactive workflow. Establishes the technical baseline every subsequent chapter depends on.

Lesson 4 • Introduction to DataFrames

Construct DataFrames from dictionaries, lists, and Series objects. Provides the foundational mental model for all data manipulation covered later.

Lesson 1 • Writing Data to Files

Export DataFrames to CSV, Excel, and JSON using to_csv, to_excel, and to_json. Completes the read-transform-write cycle introduced in this chapter.

Lesson 2 • Reading Excel and JSON Files

Import multi-sheet Excel workbooks and nested JSON structures into DataFrames. Expands ingestion skills beyond flat files to common business formats.

Lesson 3 • Reading from Databases

Connect to SQL databases and execute queries directly into DataFrames using read_sql. Bridges database workflows with Pandas-based analysis pipelines.

Lesson 4 • Reading Flat Files

Use read_csv and read_table with key parameters to import delimited files. Covers the most common real-world data ingestion scenario in this chapter.

Lesson 1 • Position-Based Indexing with iloc

Use iloc for integer-position-based row and column access. Complements loc when working with unlabeled or numerically indexed DataFrames.

Lesson 2 • Query Method and Advanced Selection

Use the query method and MultiIndex selection for expressive, readable filters. Extends selection skills to hierarchical indexes introduced later in the course.

Lesson 3 • Column and Row Selection Basics

Select single and multiple columns using bracket and dot notation. Establishes the simplest selection patterns before introducing label and position indexers.

Lesson 4 • Label-Based Indexing with loc

Use loc to select rows and columns by label, including slices and lists. Provides precise, readable selection that avoids positional ambiguity.

Lesson 1 • Renaming and Restructuring Columns

Rename columns with rename, clean column names, and reorder or drop columns. Consistent naming conventions improve code readability across the project.

Lesson 2 • Removing and Deduplicating Data

Find and remove duplicate rows using duplicated and drop_duplicates. Prevents inflated counts and skewed statistics in aggregation steps.

Lesson 3 • Replacing and Correcting Values

Use replace, map, and clip to fix erroneous or out-of-range values in columns. Ensures data integrity before statistical analysis or machine learning pipelines.

Lesson 4 • Handling Missing Values

Identify NaN locations with isna and isnull, then fill or drop them strategically. Missing-value handling directly affects the accuracy of every aggregation.

Lesson 1 • String Operations with str Accessor

Use the str accessor for vectorized string cleaning, splitting, and extraction. Text columns require these methods before they can be used in analysis.

Lesson 2 • Binning and Discretization

Convert continuous variables into categorical bins using cut and qcut. Discretization enables group-based analysis of numeric distributions.

Lesson 3 • Sorting and Ranking Data

Sort DataFrames by values or index and assign ranks to numeric columns. Ordered data is required for window functions and time-series operations ahead.

Lesson 4 • Creating and Modifying Columns

Derive new columns through arithmetic, string operations, and conditional logic. New features directly improve the analytical depth of downstream aggregations.

Lesson 1 • Built-In Aggregation Functions

Apply sum, mean, count, min, max, and std to grouped data efficiently. Built-in functions are faster than custom apply calls and cover most reporting needs.

Lesson 2 • Custom Aggregations and Transform

Write custom aggregation functions and use transform to broadcast group results back to the original index. Transform enables group-normalized feature creation.

Lesson 3 • GroupBy Fundamentals

Split a DataFrame into groups by one or more keys and inspect the resulting GroupBy object. Understanding the split-apply-combine pattern is the foundation of this chapter.

Lesson 4 • Reshaping with Pivot and Melt

Use pivot to reshape long data to wide format and melt to reverse the operation. Wide-to-long and long-to-wide transformations are essential for tidy data workflows.

Lesson 1 • Merging on Different Key Names

Use left_on and right_on to merge DataFrames with differently named key columns. Handles the common real-world case where source systems use inconsistent naming.

Lesson 2 • Concatenating DataFrames

Stack DataFrames vertically or horizontally with concat and handle index alignment. Concatenation is the simplest combination method and the starting point for this chapter.

Lesson 3 • Combining with Update and Combine

Use update and combine_first to fill gaps in one DataFrame from another. Enables incremental data refresh patterns common in production pipelines.

Lesson 4 • Index-Based Joining

Use join to combine DataFrames aligned on their indexes rather than columns. Index joins are faster when DataFrames share a meaningful shared index.

Lesson 1 • Time-Based Selection and Slicing

Select date ranges using partial string indexing and loc with datetime slices. Enables fast extraction of fiscal periods, quarters, and custom date windows.

Lesson 2 • Time Offsets and Period Arithmetic

Use DateOffset, Timedelta, and Period objects for date arithmetic and period indexing. Enables precise business-day calculations and fiscal-period alignment.

Lesson 3 • Resampling and Frequency Conversion

Aggregate time series to lower frequencies and upsample to higher ones with resample. Resampling is the core operation for building daily, weekly, and monthly reports.

Lesson 4 • Datetime Indexing and Parsing

Convert string columns to DatetimeIndex and set them as the DataFrame index. A proper DatetimeIndex unlocks all time-aware selection and resampling methods.