Several ways to draw timeline diagrams

I shared an automated production earlier《历史上的今天》Timeline pictures for articles,The feedback from friends is generally good,Especially the way to make the timeline,还是非常巧妙的.Today we are going to share a few different ways to make it,You can compare the pros and cons of each method by yourself

Matplotlib 制作

Matplotlib 作为 Python The most important visualization tool in the family,其基本的 API And the drawing process still needs to be mastered.Especially how flexible the library is and serves as the basis for many tools,重要性不言而喻

Let's take a look at how to draw a timeline chart

Import libraries and settings XY 轴数据

import matplotlib.pyplot as plt
plt.rcParams['font.sans-serif'] = ['SimHei']
plt.rcParams['axes.unicode_minus'] = False

y1 = [5, 10, 15, 20, 25, 30, 35, 40, 45, 50]
x1 = [4, 4, 4, 4, 4, 4, 4, 4, 4, 4]

Because the timeline effect is achieved through a line chart,In order to achieve a vertical line situation,这里设置了 X The axis values ​​are all the same,Y Axial values ​​are equally distributed

Create canvas and title

fig, ax = plt.subplots(sharey=True, figsize=(7, 4))
ax.plot(x1, y1, label='First line', linewidth=3, color='r', marker='o', markerfacecolor='white', markersize=12)
plt.title('萝卜大杂烩')
plt.suptitle('历史上的今天', fontsize=16, color='red')

此时效果如下

Next we set the data on both sides of the timeline

# 右侧数据
right_y_year = 0.95
right_y_text = 0.9
year_right = 1931
for i in range(5):
    plt.text(0.57, right_y_year, str(year_right+1), fontsize=15, horizontalalignment='center', verticalalignment='center', 
             transform=ax.transAxes, color='black')
    plt.text(0.75, right_y_text, "从百草园到三味书屋-鲁迅" + str(i), fontsize=15, horizontalalignment='center', verticalalignment='center', 
             transform=ax.transAxes, color='red')
    right_y_year -= 0.2
    right_y_text -= 0.2
    year_right += 1
    
# 左侧数据
left_y_year = 0.85
left_y_text = 0.8
year_left = 1941
for i in range(5):
    plt.text(0.43, left_y_year, str(year_left+1), fontsize=15, horizontalalignment='center', verticalalignment='center', 
             transform=ax.transAxes, color='black')
    plt.text(0.2, left_y_text, "从百草园到三味书屋-鲁迅" + str(i), fontsize=15, horizontalalignment='center', verticalalignment='center', 
             transform=ax.transAxes, color='red', url='https://www.baidu.com')
    left_y_year -= 0.2
    left_y_text -= 0.2
    year_left += 1

这里主要使用了 text 函数,Add data to each side of the timeline

If we also want to add other personal information,Such as public account QR code, etc,You can add pictures at specified locations,Also remove the axes

# 增加图片
img = plt.imread('二维码.png')
ax2 = plt.axes((0.7, 0.1, 0.3, 0.3))
ax2.imshow(img, origin='lower', alpha=0.5)
ax2.axis('off')
ax.axis('off')
plt.show()

最终效果如下

可以看出,由于 text The function determines the position of the text display by the coordinates,So the data distribution on both sides of our timeline is still not particularly perfect,I don't know if there is another more convenient way to set it up

Plotly 绘制

Plotly 作为 Python Another very powerful visualization tool in the family,The drawing of the timeline diagram can also be completed

在绘图之前,Perfect to process the data first

The data used here is2020Weibo hot search data throughout the year

import pandas as pd


weibo = pd.read_csv("weibo_2020.csv")
def deal_date(frame):
    tmp = frame.split('-')
    return tmp[0] + '-' + tmp[1]

weibo['new_date'] = weibo['date'].apply(lambda x : deal_date(x))
key_list_right = []
for i in range(1, 12, 2):
    if i < 10:
        mydate = '2020-0%s' % str(i)
    else:
        mydate = '2020-%s' % str(i)
    keyword = weibo[weibo['new_date'] == mydate].sort_values(by='searchCount', ascending=False)['keyword'].tolist()[0]
    searchCount = weibo[weibo['new_date'] == mydate].sort_values(by='searchCount', ascending=False)['searchCount'].tolist()[0]
    mount = str(i) + '月'
    content = ','.join([keyword, str(searchCount) + '搜索量', mount])
    key_list_right.append(content)
print(key_list_right)

Output:

['The latest epidemic map,18130201搜索量,1月',
 '肖战工作室道歉,13117531搜索量,3月',
 'Stanley Ho passed away,15302424搜索量,5月',
 '高考作文,15647446搜索量,7月',
 'Sisters who ride the wind and waves form a group night,8226994搜索量,9月',
 '特朗普,7310000搜索量,11月']

可以看到,Through the above data processing,We successfully extracted1、3、5、7、9以及11The most searched topic of the month for the month,In the same way, you can obtain the hot search title data of the two months

下面开始作图

import plotly.express as px
import plotly.graph_objects as go
from plotly.subplots import make_subplots
import pandas as pd
from plotly.graph_objs import *

layout = Layout(
    paper_bgcolor='rgba(0,0,0,0)',
    plot_bgcolor='rgba(0,0,0,0)',
    title={'text': '微博热搜', 'x': 0.5},
    yaxis={'title': 'Proportion (%)'}
)
fig = go.Figure(layout=layout)
fig.add_traces([go.Scatter(x=[2,2,2,2,2,2], y=[5, 10, 15, 20, 25, 30], text=key_list_right, textposition="bottom right", mode="lines+text"),
                go.Scatter(x=[2,2,2,2,2,2], y=[5, 10, 15, 20, 25, 30], textposition="top left", mode="lines+text", text=key_list_left)]) 
fig.update_traces(showlegend=False)
fig.update_layout(xaxis=dict(visible=False), yaxis=dict(visible=False))
fig.show()

通过 Plotly Drawing is relatively simple,直接使用 text The parameters can be added to the hot search data we get

最终效果如下

The effect is simple,It's because we didn't do too much styling,You can explore different styles by yourself

Excel 绘制

Both of the above methods require a certain code base,下面介绍的 Excel The method can be said that everyone can do it,一起来看看吧

先来看看最终的效果

首先准备数据,我们在新建的 Excel Create the following data in the document

Then insert a scatter plot

Insert a blank scatterplot first,然后将 X 轴设置为【年份】,Y 轴设置为【位置】,再把 Y Both axes and gridlines are removed

Next let's beautify it X 轴

我们双击 X 轴,调出格式窗口,Set in the Axis Options tab【单位】,将【小】改为1,设置【刻度线】,将【主刻度线】set to cross.再点击【油漆桶】,Choose a line color,Adjust the width to 2,将【End arrow type】Adjust to right arrow

Next we put X 轴连接起来

First select a scatter point,添加误差线.Then set the horizontal error bars to no outline,Select the vertical error bars again,把【Vertical error bars】Set to negative bias,Then set the error amount to 100%,Finally, adjust the style of the vertical error bars

Let's start adding data

We add the company's big events to the data table

Add to the chart【数据标签】,That is, the column of events in the data,然后再去掉 Y 值即可

最后我们还可以通过 Excel Comes with various icons for beautification operations

好了,以上就是今天分享的所有内容,如果对你有帮助,Please like and support~