6 - Data Visualization with Matplotlib and Intro to Pandas — Topics & Learning Outcomes

Module Topics

Introduction to Matplotlib and the pyplot Interface

• What is Matplotlib? — Matplotlib is a comprehensive Python library for creating static, animated, and interactive visualizations. It is the foundational plotting library in the scientific Python ecosystem.
• The pyplot Module — The pyplot module provides a state-based interface to matplotlib, allowing users to create and modify figures with simple function calls. It is typically imported as plt.
• Figures and Axes Objects — In matplotlib, a Figure is the top-level container for all plot elements, while an Axes is the region within the figure where data is plotted. Understanding this hierarchy is essential for building complex visualizations.
• Displaying and Saving Figures — Once a plot is constructed, it can be rendered to the screen with plt.show() or saved to a file with plt.savefig(). Proper use of these functions ensures figures appear and are preserved correctly.

Line Plots and Scatter Plots

• Creating a Basic Line Plot — A line plot is created using plt.plot(x, y), which connects data points with line segments. It is ideal for showing trends over a continuous variable such as time or distance.
• Customizing Line Plots — Line plots can be customized with keyword arguments to control color, line width, style, and markers. Meaningful labels and titles improve interpretability.
• Creating a Scatter Plot — A scatter plot is created with plt.scatter(x, y) and displays individual data points without connecting lines. It is useful for exploring correlations and distributions between two variables.
• Axis Limits and Scaling — Controlling the visible range of axes ensures that important data features are not clipped or obscured. Matplotlib provides functions to set both linear and logarithmic scales.

Bar Charts and Histograms

• Creating a Bar Chart — Bar charts are created with plt.bar(categories, values) and are used to compare discrete groups or categories. They are commonly used to display experimental results across different conditions.
• Annotating Bar Charts — Adding value labels or annotations to bar charts makes them easier to interpret, especially in technical reports. matplotlib provides text placement functions for this purpose.
• Creating a Histogram — Histograms are created with plt.hist(data, bins=n) and display the frequency distribution of a single continuous variable. They are essential for understanding data spread and identifying outliers in engineering measurements.
• Comparing Distributions with Histograms — Multiple datasets can be compared by overlaying histograms on the same axes. Proper use of transparency and color enables effective visual comparison of distributions.

Subplots and Figure Layout

• Creating Subplots with plt.subplots() — The plt.subplots(nrows, ncols) function creates a Figure and a grid of Axes objects in one call. It is the recommended approach for building multi-panel layouts.
• Plotting on Individual Axes — When using subplots, plotting commands are called on individual Axes objects rather than on plt directly. This object-oriented approach provides precise control over each panel.
• Adjusting Layout and Spacing — Subplot panels can overlap or clip labels if spacing is not adjusted. matplotlib provides tools to automatically or manually correct figure layout.
• Using GridSpec for Complex Layouts — For non-uniform subplot grids, matplotlib.gridspec.GridSpec allows plots of different sizes to share a figure. This is useful for dashboard-style engineering figures.

Publication-Quality Figure Styling

• Using Style Sheets — Matplotlib includes built-in style sheets that instantly apply consistent aesthetics to all figures in a script. Selecting an appropriate style reduces manual customization effort.
• Customizing Colors and Colormaps — Color choices affect both the aesthetic quality and accessibility of a figure. Matplotlib supports named colors, hex codes, RGB tuples, and scientific colormaps.
• Fonts, Tick Marks, and Spines — Fine-tuning typography and axis decorations distinguishes a presentation-ready figure from a draft plot. Matplotlib allows control over font size, tick frequency, and axis border (spine) visibility.
• Saving High-Resolution Figures — Exporting figures at sufficient resolution is critical for print publications and technical reports. The plt.savefig() function supports multiple formats and resolution settings.

Introduction to Pandas: Series and DataFrames

• What is Pandas? — Pandas is a Python library built on NumPy that provides high-performance, easy-to-use data structures for working with labeled and relational data. It is the standard tool for tabular data analysis in Python.
• The Pandas Series — A Series is a one-dimensional labeled array capable of holding data of any type. It behaves like a NumPy array but with an associated index that labels each element.
• The Pandas DataFrame — A DataFrame is a two-dimensional labeled data structure with columns of potentially different types, analogous to a spreadsheet or SQL table. It is the primary pandas object for engineering datasets.
• Inspecting a DataFrame — Pandas provides several methods for quickly exploring the structure and content of a newly loaded dataset. These are essential first steps in any data analysis workflow.

Reading, Filtering, and Sorting Data with Pandas

• Reading Data from Files — Pandas provides read functions that import data from common file formats directly into a DataFrame. Reading CSV files is the most common operation in engineering data workflows.
• Selecting Columns and Rows — Pandas provides multiple methods for selecting subsets of a DataFrame by column name or row index. Mastery of these selection methods is fundamental to data manipulation.
• Filtering Rows with Boolean Conditions — Boolean indexing allows rows to be selected based on conditions applied to one or more columns. This is the standard method for subsetting engineering data by threshold or category.
• Sorting DataFrames — Sorting a DataFrame rearranges rows based on the values in one or more columns. Sorted data is easier to inspect and is required as a preprocessing step for some analyses.
• Adding and Modifying Columns — New columns can be added to a DataFrame through direct assignment, enabling derived variables to be computed from existing data. This is essential for feature engineering and unit conversions.

Descriptive Statistics with Pandas

• Summary Statistics with describe() — The df.describe() method computes count, mean, standard deviation, min, max, and quartiles for all numeric columns simultaneously. It provides a rapid overview of dataset characteristics.
• Aggregation with groupby() — The groupby() method splits a DataFrame into groups based on column values and applies aggregate functions to each group. It is the pandas equivalent of SQL GROUP BY.
• Handling Missing Values — Missing values represented as NaN are common in real engineering datasets from sensor dropouts or incomplete records. Pandas provides methods to detect, remove, or fill missing data.
• Correlation and Relationships — Pandas can compute pairwise correlations between numeric columns, providing a quick measure of linear relationships within an engineering dataset.
• Plotting Directly from Pandas — Pandas DataFrames and Series have built-in plot methods that call matplotlib internally, enabling quick exploratory visualizations with minimal code.

Student Learning Outcomes

By the end of this module, students will be able to:

Course Outcomes (reference)