Visual common drawing (I) stacking diagram

One . stacked column chart （Stacked Bar Chart）

1.1 Definition

Stacked histogram can vividly show the data of each small category contained in a large category , And the proportion of each sub category , It shows the relationship between a single project and the whole . There are two types of stacked histograms ：

• General stacked histogram ： The values on each column represent different data sizes , The sum of the data of each floor represents the height of the whole column . It is very suitable for comparing the total amount of data in each group .
• Stacked histogram of percentage ： Each layer of the column represents the percentage of the category data in the overall data of the group .

1.2 Use scenarios

Applicable scenario ：

1. Apply to need Compare the total size of different groups , At the same time, compare the size of different classifications in the same group .

Not applicable to the scene ：

1. Compare the data size of the same category in different groups （ Percentage stacked histogram ）
2. Compare the size of the total number of groups （ Backup is better than stacking histogram ）

1.3 Realization

stay matplotlib Use in bar Function to implement , Appoint bottom Stack parameters , The code is as follows ：

import numpy as np
import matplotlib.pyplot as plt

#  Data set used 
year = [1950, 1960, 1970, 1980, 1990, 2000, 2010, 2018]
population_by_continent = {
    
    'africa': [228, 284, 365, 477, 631, 814, 1044, 1275],
    'americas': [340, 425, 519, 619, 727, 840, 943, 1006],
    'asia': [1394, 1686, 2120, 2625, 3202, 3714, 4169, 4560],
    'europe': [220, 253, 276, 295, 310, 303, 294, 293],
    'oceania': [12, 15, 19, 22, 26, 31, 36, 39],
}

#  initialization figure and axes
fig, ax = plt.subplots()
sum = np.zeros((1, len(year))).reshape(-1)

#  mapping 
for i in population_by_continent:
    ax.bar(year, population_by_continent.get(i), label=i, alpha=0.8, bottom=sum, width=5)
    sum += population_by_continent.get(i)

#  Add legend and title 
ax.legend(loc='upper left')
ax.set_title('World population')
ax.set_xlabel('Year')
ax.set_ylabel('Number of people (millions)')

plt.show()

The effect is as follows ：

---

Two . Stacking area map （Stacked Area Chart）

2.1 Definition

Stacked area map draws multiple data sets as vertically stacked areas . The stacked area map is the same as the basic area map , The only difference is that the starting point of each data set on the graph is different , The starting point is based on the previous dataset , Used to display the trend line of the proportion of each value over time or category , It shows the relationship between part and whole .

The largest area on the stacked area graph represents the sum of all data volumes , It's a whole . Each stacked area represents the size of each data volume , These stacked area maps are particularly useful in showing the changes of the total component of big data .

Based on the stacked area map , The data of the dependent variables of each area are normalized by the total amount after addition to form a percentage stacked area map . The chart does not reflect the change of the total amount , But you can clearly

2.2 Use scenarios

Applicable scenario ：

1. Applicable to trend lines that need to reflect the percentage of each value over time or category , For analysis, the independent variable is time-varying data 、 The proportion of each index component is very useful when using ordered data .
2. It is suitable for comparing multivariable changes over time .

Not applicable to the scene ：

1. Not applicable with negative Data set of .

2.3 Realization

Use stackplots Realize the stacking area diagram ,stackplots The function parameters are as follows ：

effect ： Draw stacking area map

 stackplot(x, *args, labels=(), colors=None, baseline='zero', data=None, **kwargs)

Parameters 1：x：$(N,)$ Array , Specify the horizontal axis .

Parameters 2：*args,y：$(N,M)$ Array , Specify the data to plot .

Parameters 3：labels： String list ： Specify the label for each drawing data .

Parameters 4：colors： Color list ： Specifies the shading of the stacked area , Not necessarily with y Same end , Colors will be recycled .

Parameters 5：baseline： Qualified string ： Specify the method of calculating the baseline , Optional ：

• zero： Constant zero baseline , A simple stacked graph .
• sym： Symmetry around zero
• wiggle： Minimize the sum of the square slopes .
• weighted_wiggle: Weighted , The weight accounts for the size of each layer

Parameters 6：data： Unknown

Parameters 7：**kwargs： The received keyword parameters are passed to fill_between Method .

import numpy as np
import matplotlib.pyplot as plt


year = [1950, 1960, 1970, 1980, 1990, 2000, 2010, 2018]
population_by_continent = {
    
    'africa': [228, 284, 365, 477, 631, 814, 1044, 1275],
    'americas': [340, 425, 519, 619, 727, 840, 943, 1006],
    'asia': [1394, 1686, 2120, 2625, 3202, 3714, 4169, 4560],
    'europe': [220, 253, 276, 295, 310, 303, 294, 293],
    'oceania': [12, 15, 19, 22, 26, 31, 36, 39],
}

fig, ax = plt.subplots()
ax.stackplot(year, population_by_continent.values(),
             labels=population_by_continent.keys(), alpha=0.8)
ax.legend(loc='upper left')
ax.set_title('World population')
ax.set_xlabel('Year')
ax.set_ylabel('Number of people (millions)')

plt.show()

The effect is as follows ：