Plotting graph using Seaborn | Python

Seaborn is a Python data visualization library built on top of Matplotlib. It provides a high-level interface for drawing attractive, informative statistical graphics. Unlike Matplotlib, Seaborn works seamlessly with Pandas DataFrames, making it a preferred tool for quick exploratory data analysis and advanced statistical plotting.

• Comes with built-in datasets like iris, tips, etc.
• Provides statistical plots such as boxplots, violin plots, swarm plots, etc.
• Handles categorical data visualization better than Matplotlib.
• Supports aesthetic customization (themes, color palettes, styles).
• Simplifies working with DataFrames by auto-labeling axes.

Different Plots in Seaborn

Let's see the various types of plots in seaborn,

1. Strip Plot

A strip plot is a categorical scatter plot where data points are plotted along one categorical axis. It is useful for visualizing the distribution of values but may suffer from overlapping points.

Applications

• Used when we want to visualize raw distribution of numerical data across categories.
• Helpful for detecting clusters or general spread of values.

Advantages

• Simple and easy to interpret.
• Shows individual data points clearly.

Limitations

• Overlapping points may cause loss of clarity in dense datasets.

Output:

2. Swarm Plot

A swarm plot is similar to a strip plot, but points are arranged to avoid overlap. This ensures all data points are visible, making it more informative.

Applications

• Useful when dataset is small/medium and we want to show all observations.
• Comparing sub-groups clearly without stacking.

Advantages

• Prevents overlap of data points.
• Provides clearer visual insight than strip plot.

Limitations

• Can be slow for large datasets.
• May look cluttered when categories have thousands of points.

Output:

3. Bar Plot

A bar plot shows the average (by default mean) of a numerical variable across categories. It can use different estimators (mean, median, std, etc.) for aggregation.

Applications

• Comparing average values across categories.
• Displaying results of group-by operations visually.

Advantages

• Easy to interpret and widely used.
• Flexible can use different statistical functions.

Limitations

• Does not show individual data distribution.
• Can hide variability when using only mean.

Output:

4. Count Plot

A count plot simply counts the occurrences of each category. It is like a histogram for categorical variables.

Applications

• Checking frequency distribution of categorical values.
• Understanding class imbalance in data.

Advantages

• Very simple and quick to interpret.
• No need for numerical data, only categorical required.

Limitations

• Cannot display numerical spread inside categories.

Output:

5. Box Plot

A box plot (or whisker plot) summarizes numerical data using quartiles, median and outliers. It helps in detecting variability and spread.

Applications

• Detecting outliers.
• Comparing spread of distributions across categories.

Advantages

• Highlights summary statistics effectively.
• Useful for large datasets.

Limitations

• Does not show exact data distribution shape.

Output:

6. Violin Plot

A violin plot combines a box plot with a density plot, showing both summary stats and distribution shape.

Applications

• Comparing distributions more deeply than boxplot.
• Helpful for detecting multimodal distributions.

Advantages

• Shows both summary statistics and data distribution.
• Easier to see differences in distribution shapes.

Limitations

• Can be harder to interpret for beginners.
• May be misleading if sample size is small.

Output:

7. Strip Plot with Hue

This is an enhanced strip plot where categories are further divided using hue. It allows comparing multiple sub-groups within a category.

Applications

• Comparing subgroups inside categories.
• Visualizing interaction between two categorical variables.

Advantages

• Adds extra dimension to strip plot.
• Useful for multivariate visualization.

Limitations

• Overlap issue exists.

Output:

Applications

• Exploratory Data Analysis (EDA): Identifying trends, outliers and patterns.
• Feature Analysis: Comparing numerical features across categories.
• Data Presentation: Creating professional, publication-ready plots.
• Model Preparation: Checking class imbalance or spread before training models.