Seaborn Scatterplot

Prerequisite and How to Code Along

It's recommended to use Google Colab or Jupyter Notebook or Jupyter Lab in Anaconda.

Importing The Libraries

import matplotlib.pyplot as plt
import seaborn as sns
import pandas as pd

Loading Data

We will try to load data from various types like list, tuple, dict or dataframe.

Loading Data From List or Tuple

horizontal = [1, 2, 3, 4, 5, 2, 4, 3]
vertical = [2, 4, 6, 8, 10, 6, 4, 9]

sns.scatterplot(x=horizontal, y=vertical)

horizontal = (1, 2, 3, 4, 5, 2, 4, 3)
vertical = (2, 4, 6, 8, 10, 6, 4, 9)

sns.scatterplot(x=horizontal, y=vertical)

Loading Data From Dictionary

dict_data = {
    'horizontal': (1, 2, 3, 4, 5, 2, 4, 3),
    'vertical': (2, 4, 6, 8, 10, 6, 4, 9),
}

sns.scatterplot(x='horizontal', y='vertical', data=dict_data)

Loading Data From Pandas DataFrame

dict_data = {
    'horizontal': [1, 2, 3, 4, 5, 2, 4, 3],
    'vertical': [2, 4, 6, 8, 10, 6, 4, 9],
}

df = pd.DataFrame(dict_data)

sns.scatterplot(x='horizontal', y='vertical', data=df)

Any of the above code will display the same plot like this:

Styling

Dots Color

Using Web Color Name:

sns.scatterplot(x='horizontal', y='vertical', data=df, c=['red'])

or Using Hex

sns.scatterplot(x='horizontal', y='vertical', data=df, c=['#c21b95'])

Dots Transparency

sns.scatterplot(
    x='horizontal',
    y='vertical',
    data=df,
    c=['#c21b95'],
    alpha=0.4
)

Hue

We can classify each dot depends on another feature. We add new feature named class that has category name for each dots.

df = pd.DataFrame({
    'horizontal': [1, 2, 3, 4, 5, 2, 4, 3],
    'vertical': [2, 4, 6, 8, 10, 6, 4, 9],
    'class': ['category 2', 'category 1', 'category 3', 'category 2', 'category 3', 'category 1', 'category 3', 'category 2']
})

sns.scatterplot(x='horizontal', y='vertical', data=df, hue='class', hue_order=['category 3', 'category 1', 'category 2'])

Blog

Dendi Handian