This repository contains code about creating interesting plots in Python.

1. Loading the data

First we need to load\create the data:

import pandas as pd
import numpy as np

np.random.seed(0)  # For reproducibility

data = {
    'Normal distribution': np.random.normal(loc=50, scale=10, size=100), 
    'Exponential distribution': np.random.exponential(scale=1, size=100),     
    'Uniform distribution': np.random.uniform(low=20, high=80, size=100), 
    'Poisson distribution': np.random.poisson(lam=3, size=100)            
}

# Creating a dataframe
df = pd.DataFrame(data)

2. Creating the plots

Next, we create the plots:

• Histogram

The plots can be overlaying:

import matplotlib.pyplot as plt

plt.rcParams.update({"font.family": "Bell MT"}) #you can specify the font of the texts on the plots
plt.rcParams.update({'font.size': 14}) #you can specify the font size of the texts on the plots, Python will use this font all over the code, unless said otherwise.

plt.hist(df['Normal distribution'], bins=30, color='limegreen', edgecolor='black', density=True, label='Normal Distribution', alpha=0.7) #alpha is the transparency
plt.hist(df['Exponential distribution' ], bins=30, color='skyblue', edgecolor='black', density=True, label='Exponential Distribution', alpha=0.6)
plt.hist(df['Uniform distribution'], bins=30, color='pink', edgecolor='black', density=True, label='Uniform Distribution', alpha=0.5)
plt.hist(df['Poisson distribution' ], bins=30, color='orange', edgecolor='black', density=True, label='Poisson Distribution', alpha=0.6)
plt.legend()
plt.xlabel('Values')
plt.ylabel('Probability Density')
plt.title('Histogram of Different Distributions')
#plt.savefig('*.png', dpi=300) #you can also save your plot
plt.show()

‎‎

You can find the lists of colors in python here. Moreover, the color codes for Python can be found here.

The histograms can be shown in the form of subplots in a plot:

import matplotlib.cm as cm

# Plotting histograms in the form og subplots
fig, axes = plt.subplots(nrows=2, ncols=2, figsize=(12, 10))

# Column names
columns = df.columns

# Define a colormap to show each subplot with a distinct color
cmap = cm.get_cmap('viridis', len(df.columns))

# Plotting each histogram in a subplot
for i, ax in enumerate(axes.flatten()):
    color = cmap(i)  # Get a distinct color from the colormap
    ax.hist(df[columns[i]], bins=30, color=color, edgecolor='black', density=True, alpha=0.7)
    ax.set_title(f'Histogram of {columns[i]}')
    ax.set_xlabel('Values')
    ax.set_ylabel('Probability Density')

# Adjust layout
plt.tight_layout()

# Display the plot
plt.show()

Or we canhave the histograms to be shown seperately, each in its own subplot.

# Define a colormap
cmap = cm.get_cmap('viridis', len(df.columns))

# Plotting histograms for each column
for i,column in enumerate(df.columns):
    color = cmap(i)
    plt.figure(figsize=(8, 6))
    plt.hist(df[column], bins=30, color=color, edgecolor='black', density=True, alpha=0.7)
    plt.title(f'Histogram of {column}')
    plt.xlabel('Values')
    plt.ylabel('Probability Density')
    plt.show()

• Boxplots

import matplotlib.pyplot as plt

plt.rcParams.update({"font.family": "Bell MT"})
plt.rcParams.update({'font.size': 14}) 

fig = plt.figure(figsize =(10, 7))
ax = fig.add_subplot(111)
 
# Creating axes instance
bp = ax.boxplot(df, patch_artist = True,
                notch ='True', vert = 0) #notch is the way the plots are narrower in the middle
                                         # vert is the option for showing the boxplots vertically or horizontally
colors = ['#0000FF', '#00FF00', 
          '#FFFF00', '#FF00FF'] #the colors of the plots
 
for patch, color in zip(bp['boxes'], colors):
    patch.set_facecolor(color)
 
# changing color and linewidth of whiskers
for whisker in bp['whiskers']:
    whisker.set(color ='#8B008B',
                linewidth = 1.5,
                linestyle =":")
 
# changing color and linewidth of caps
for cap in bp['caps']:
    cap.set(color ='#8B008B',
            linewidth = 2)
 
# changing color and linewidth of medians
for median in bp['medians']:
    median.set(color ='red',
               linewidth = 3)
 
# changing style of fliers
for flier in bp['fliers']:
    flier.set(marker ='s', 
              color ='#e7298a',
              alpha = 0.5)
     
ax.set_yticklabels(['Normal', 'Exponential', 
                    'Uniform', 'Poisson'])
 
plt.title("Distributions Boxplots")
 
ax.get_xaxis().tick_bottom()
ax.get_yaxis().tick_left()
plt.show()

The source of the above code is this website.

Moreover, we can show them in different subplots of the same plot:

# Create a figure with subplots
fig, axes = plt.subplots(4, 1, figsize=(10, 14))  # 4 rows, 1 column

# Colors for the boxplots
colors = ['#0000FF', '#00FF00', '#FFFF00', '#FF00FF']

# Plot each distribution in a separate subplot
for ax, (column, color) in zip(axes, zip(df.columns, colors)):
    bp = ax.boxplot(df[column], patch_artist=True, notch=True, vert=0)

    # Set color for boxes
    for patch in bp['boxes']:
        patch.set_facecolor(color)

    # Set color and linewidth for whiskers
    for whisker in bp['whiskers']:
        whisker.set(color='#8B008B', linewidth=1.5, linestyle=":")

    # Set color and linewidth for caps
    for cap in bp['caps']:
        cap.set(color='#8B008B', linewidth=2)

    # Set color and linewidth for medians
    for median in bp['medians']:
        median.set(color='red', linewidth=3)

    # Set style for fliers
    for flier in bp['fliers']:
        flier.set(marker='s', color='#e7298a', alpha=0.5)
    
    # Set title and labels
    ax.set_title(f"{column} Boxplot")
    ax.get_xaxis().tick_bottom()
    ax.get_yaxis().tick_left()
    ax.set_yticklabels([column.split()[0]])

# Set overall title
fig.suptitle("Distributions Boxplots", fontsize=16)

# Adjust layout
plt.tight_layout(rect=[0, 0, 1, 0.96])
plt.show()

Moreover, boxplots can be generate by seaborn package as well. For more information on the basic of this package please refer to this link.

• Violin Boxplots

This kind of plot is used to show both the boxplot and the distribution of a dataset.

# Colors for the violin plots
my_pal = {'limegreen',  'skyblue', 'pink', 'orange'}

ax = sns.violinplot(data=[df['Normal distribution'], df['Exponential distribution'],
                          df['Uniform distribution'], df['Poisson distribution']], 
                    palette=my_pal, inner=None)

#you can apply transparecy to this plot
# for patch in ax.collections:
#     patch.set_alpha(0.3) 

# Adjust violin plot width
for item in ax.collections:
    x0, y0, width, height = item.get_paths()[0].get_extents().bounds
    item.set_clip_path(plt.Rectangle((x0, y0), width/2, height,
                       transform=ax.transData))

# Adjust violin plot offsets
num_items = len(ax.collections)
for item in ax.collections[num_items:]:
    item.set_offsets(item.get_offsets() + 0.15)

# Create boxplot
sns.boxplot(data=[df['Normal distribution'], df['Exponential distribution'],
                          df['Uniform distribution'], df['Poisson distribution']], width=0.25,
            showfliers=True, showmeans=True, 
            meanprops=dict(marker='o', markerfacecolor='maroon',
                           markersize=10, zorder=1),
            boxprops=dict(facecolor=(0,0,0,0), 
                          linewidth=1, zorder=3),
            whiskerprops=dict(linewidth=1),
            capprops=dict(linewidth=1),
            medianprops=dict(linewidth=1))

# Set labels and ticks
ax.set_xticklabels(['Normal\ndistribution', 'Exponential\ndistribution',
                    'Uniform\ndistribution','Poisson\ndistribution'], size=14, fontweight='bold')
ax.set_ylabel("Values", size=16, fontweight='bold')
plt.yticks(fontweight='bold', size=14)
plt.title ('Distributions Violin Boxplots', fontweight='bold')

plt.show()

Or show then in different subplots:

# Create a figure with subplots
fig, axes = plt.subplots(2, 2, figsize=(14, 10))  # 2 rows, 2 columns

# Colors for the violin plots
my_pal = ['limegreen', 'skyblue', 'pink', 'orange']

# List of distributions
distributions = ['Normal distribution', 'Exponential distribution', 'Uniform distribution', 'Poisson distribution']

# Plot each distribution in a separate subplot
for ax, column, color in zip(axes.flatten(), distributions, my_pal):
    # Violin plot
    sns.violinplot(ax=ax, data=df[column], color=color, inner=None)
    
    # Adjust violin plot width
    for item in ax.collections:
        x0, y0, width, height = item.get_paths()[0].get_extents().bounds
        item.set_clip_path(plt.Rectangle((x0, y0), width/2, height,
                           transform=ax.transData))

    # Boxplot
    sns.boxplot(ax=ax, data=df[column], width=0.25,
                showfliers=True, showmeans=True, 
                meanprops=dict(marker='o', markerfacecolor='maroon',
                               markersize=10, zorder=1),
                boxprops=dict(facecolor=(0,0,0,0), 
                              linewidth=1, zorder=3),
                whiskerprops=dict(linewidth=1),
                capprops=dict(linewidth=1),
                medianprops=dict(linewidth=1))
    
    # Set labels and ticks
    ax.set_xticklabels([f'{column.split()[0]}\n{column.split()[1]}'], size=14, fontweight='bold')
    ax.set_ylabel("Values", size=16, fontweight='bold')

# Set overall title
fig.suptitle("Distributions Violin Boxplots", fontsize=16, fontweight='bold')

# Adjust layout
plt.tight_layout(rect=[0, 0, 1, 0.96])
plt.show()

• Scatter Plot

Plotting the distributions in different subplots as a scatter plot:

# Plotting
fig, axs = plt.subplots(2, 2, figsize=(12, 8))

# Define colormaps
colormaps = ['viridis', 'plasma', 'inferno', 'cividis']

# Plot each distribution with a different colormap
for (ax, distribution, colormap) in zip(axs.flatten(), df.columns, colormaps):
    sc = ax.scatter(range(len(df)), df[distribution], c=df[distribution], cmap=colormap)
    ax.set_title(distribution)
    ax.set_xlabel('Index')
    ax.set_ylabel('Value')
    ax.grid(True)
    ax.set_facecolor('whitesmoke')
    ax.set_axisbelow(True)
    fig.colorbar(sc, ax=ax, orientation='vertical', pad=0.05)

# Adjust layout
plt.tight_layout()
plt.show()