作者：白寧超

2017年7月19日09:09:07

摘要：資料視覺化主要旨在藉助於圖形化手段，清晰有效地傳達與溝通訊息。但是，這並不就意味著資料視覺化就一定因為要實現其功能用途而令人感到枯燥乏味，或者是為了看上去絢麗多彩而顯得極端複雜。為了有效地傳達思想概念，美學形式與功能需要齊頭並進，通過直觀地傳達關鍵的方面與特徵，從而實現對於相當稀疏而又複雜的資料集的深入洞察。然而，設計人員往往並不能很好地把握設計與功能之間的平衡，從而創造出華而不實的資料視覺化形式，無法達到其主要目的，也就是傳達與溝通訊息。資料視覺化與資訊圖形、資訊視覺化、科學視覺化以及統計圖形密切相關。當前，在研究、教學和開發領域，資料視覺化乃是一個極為活躍而又關鍵的方面。“資料視覺化”這條術語實現了成熟的科學視覺化領域與較年輕的資訊視覺化領域的統一。（本文原創編著，轉載註明出處

:資料視覺化例項分析）

1 折線圖的製作

1.1 需求描述

使用matplotlib繪製一個簡單的折線圖，在對其進行定製，以實現資訊更加豐富的資料視覺化，繪製（1,2,3,4,5）的平方折線圖。

1.2 原始碼

#coding=utf-8
import matplotlib as mpl
import matplotlib.pyplot as plt
import pylab
# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

 
# squares = [1,35,43,3,56,7]
input_values = [1,2,3,4,5]
squares = [1,4,9,16,25]
# 設定折線粗細
plt.plot(input_values,squares,linewidth=5)

# 設定標題和座標軸
plt.title('平方數圖',fontsize=24)
plt.xlabel('值',fontsize=14)
plt.ylabel('平方值',fontsize=14)

# 設定刻度大小
plt.tick_params(axis='both',labelsize=14)

plt.show()
 

1.3 生成結果

2 scatter()繪製散點圖

2.1 需求描述

 使用matplotlib繪製一個簡單的雜湊點圖，在對其進行定製，以實現資訊更加豐富的資料視覺化，繪製（1,2,3,4,5）的散點圖。 

2.2 原始碼

#coding=utf-8
import matplotlib as mpl
import matplotlib.pyplot as plt
import pylab

# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

# 設定雜湊點縱橫座標值
x_values = [1,2,3,4,5]
y_values = [1,4,9,16,25]

# s設定雜湊點的大小，edgecolor='none'為刪除資料點的輪廓
plt.scatter(x_values,y_values,c='red',edgecolor='none',s=40)

# 設定標題和座標軸
plt.title('平方數圖',fontsize=24)
plt.xlabel('值',fontsize=14)
plt.ylabel('平方值',fontsize=14)

# 設定刻度大小
plt.tick_params(axis='both',which='major',labelsize=14)

# 自動儲存圖表，引數2是剪裁掉多餘空白區域
plt.savefig('squares_plot.png',bbox_inches='tight')

plt.show()

2.3 生成結果

 

2.4  需求改進

 使用matplotlib繪製一個簡單的雜湊點圖，在對其進行定製，以實現資訊更加豐富的資料視覺化，繪製1000個數的散點圖。並自動統計資料的平方，自定義座標軸

2.5  原始碼改進

#coding=utf-8
import matplotlib as mpl
import matplotlib.pyplot as plt
import pylab

# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

# 設定雜湊點縱橫座標值
# x_values = [1,2,3,4,5]
# y_values = [1,4,9,16,25]

# 自動計算資料
x_values = list(range(1,1001))
y_values = [x**2 for x in x_values]

# s設定雜湊點的大小，edgecolor='none'為刪除資料點的輪廓
# plt.scatter(x_values,y_values,c='red',edgecolor='none',s=40)

# 自定義顏色c=(0,0.8,0.8)紅綠藍
# plt.scatter(x_values,y_values,c=(0,0.8,0.8),edgecolor='none',s=40)

# 設定顏色隨y值變化而漸變
plt.scatter(x_values,y_values,c=y_values,cmap=plt.cm.Reds,edgecolor='none',s=40)

# 設定標題和座標軸
plt.title('平方數圖',fontsize=24)
plt.xlabel('值',fontsize=14)
plt.ylabel('平方值',fontsize=14)

#設定座標軸的取值範圍
plt.axis([0,1100,0,1100000])

# 設定刻度大小
plt.tick_params(axis='both',which='major',labelsize=14)

# 自動儲存圖表，引數2是剪裁掉多餘空白區域
plt.savefig('squares_plot.png',bbox_inches='tight')

plt.show()

2.6 改進結果

3 隨機漫步圖

3.1 需求描述

隨機漫步是每次步行方向和步長都是隨機的，沒有明確的方向，結果由一系列隨機決策決定的。本例項中random_walk決策步行的左右上下方向和步長的隨機性，rw_visual是圖形化展示。

3.2 原始碼

random_walk.py

from random import choice

class RandomWalk():
    '''一個生成隨機漫步資料的類'''
    def __init__(self,num_points=5000):
        '''初始化隨機漫步屬性'''
        self.num_points = num_points
        self.x_values = [0]
        self.y_values = [0]

    def fill_walk(self):
        '''計算隨機漫步包含的所有點'''
        while len(self.x_values)<self.num_points:
            # 決定前進方向及沿著該方向前進的距離
            x_direction = choice([1,-1])
            x_distance = choice([0,1,2,3,4])
            x_step = x_direction*x_distance

            y_direction = choice([1,-1])
            y_distance = choice([0,1,2,3,4])
            y_step = y_direction*y_distance

            # 拒絕原地踏步
            if x_step == 0 and y_step == 0:
                continue

            # 計算下一個點的x和y
            next_x = self.x_values[-1] + x_step
            next_y = self.y_values[-1] + y_step

            self.x_values.append(next_x)
            self.y_values.append(next_y)

rw_visual.py

#-*- coding: utf-8 -*-
#coding=utf-8
import matplotlib as mpl
import matplotlib.pyplot as plt
import pylab
from random_walk  import RandomWalk

# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

# 建立RandomWalk例項
rw = RandomWalk()
rw.fill_walk()

plt.figure(figsize=(10,6))

point_numbers = list(range(rw.num_points))

# 隨著點數的增加漸變深紅色
plt.scatter(rw.x_values,rw.y_values,c=point_numbers,cmap=plt.cm.Reds,edgecolors='none',s=1)

# 設定起始點和終點顏色
plt.scatter(0,0,c='green',edgecolors='none',s=100)
plt.scatter(rw.x_values[-1],rw.y_values[-1],c='blue',edgecolors='none',s=100)

# 設定標題和縱橫座標
plt.title('隨機漫步圖',fontsize=24)
plt.xlabel('左右步數',fontsize=14)
plt.ylabel('上下步數',fontsize=14)

# 隱藏座標軸
plt.axes().get_xaxis().set_visible(False)
plt.axes().get_yaxis().set_visible(False)

plt.show()

3.3 生成結果

 

4 Pygal模擬擲骰子

4.1 需求描述

對擲骰子的結果進行分析，生成一個擲篩子的結果資料集並根據結果繪製出一個圖形。

4.2 原始碼

 Die類

import random

class Die:
    """
    一個骰子類
    """
    def __init__(self, num_sides=6):
        self.num_sides = num_sides

    def roll(self):
        # 返回一個1和篩子面數之間的隨機數
        return random.randint(1, self.num_sides)

die_visual.py

#coding=utf-8
from die import Die
import pygal
import matplotlib as mpl
# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False


die1 = Die()
die2 = Die()
results = []
for roll_num in range(1000):
    result =die1.roll()+die2.roll()
    results.append(result)
# print(results)

# 分析結果
frequencies = []
max_result = die1.num_sides+die2.num_sides
for value in range(2,max_result+1):
    frequency = results.count(value)
    frequencies.append(frequency)
print(frequencies)

# 直方圖
hist = pygal.Bar()

hist.title = '骰子投擲1000次各面結果統計圖'
hist.x_labels =[x for x in range(2,max_result+1)]
hist.x_title ='結果'
hist.y_title = '結果分佈'

hist.add('D6+D6',frequencies)
hist.render_to_file('die_visual.svg')
# hist.show()

4.3 生成結果

 

5 同時擲兩個骰子

5.1 需求描述

對同時擲兩個骰子的結果進行分析，生成一個擲篩子的結果資料集並根據結果繪製出一個圖形。

5.2 原始碼

#conding=utf-8
from die import Die
import pygal
import matplotlib as mpl
# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

die1 = Die()
die2 = Die(10)

results = []
for roll_num in range(5000):
    result = die1.roll() + die2.roll()
    results.append(result)
# print(results)


# 分析結果
frequencies = []
max_result = die1.num_sides+die2.num_sides
for value in range(2,max_result+1):
    frequency = results.count(value)
    frequencies.append(frequency)
# print(frequencies)


hist = pygal.Bar()
hist.title = 'D6 和 D10 骰子5000次投擲的結果直方圖'
# hist.x_labels=['2','3','4','5','6','7','8','9','10','11','12','13','14','15','16']
hist.x_labels=[x for x in range(2,max_result+1)]
hist.x_title = 'Result'
hist.y_title ='Frequency of Result'

hist.add('D6 + D10',frequencies)
hist.render_to_file('dice_visual.svg')

5. 生成結果

 

6 繪製氣溫圖表

6.1 需求描述

對csv檔案進行處理，提取並讀取天氣資料，繪製氣溫表，在圖表中新增日期並繪製最高氣溫和最低氣溫的折線圖，並對氣溫區域進行著色。

6.2 原始碼

csv檔案中2014年7月部分資料資訊

AKDT,Max TemperatureF,Mean TemperatureF,Min TemperatureF,Max Dew PointF,MeanDew PointF,Min DewpointF,Max Humidity, Mean Humidity, Min Humidity, Max Sea Level PressureIn, Mean Sea Level PressureIn, Min Sea Level PressureIn, Max VisibilityMiles, Mean VisibilityMiles, Min VisibilityMiles, Max Wind SpeedMPH, Mean Wind SpeedMPH, Max Gust SpeedMPH,PrecipitationIn, CloudCover, Events, WindDirDegrees
2014-7-1,64,56,50,53,51,48,96,83,58,30.19,30.00,29.79,10,10,10,7,4,,0.00,7,,337
2014-7-2,71,62,55,55,52,46,96,80,51,29.81,29.75,29.66,10,9,2,13,5,,0.14,7,Rain,327
2014-7-3,64,58,53,55,53,51,97,85,72,29.88,29.86,29.81,10,10,8,15,4,,0.01,6,,258
2014-7-4,59,56,52,52,51,50,96,88,75,29.91,29.89,29.87,10,9,2,9,2,,0.07,7,Rain,255
2014-7-5,69,59,50,52,50,46,96,72,49,29.88,29.82,29.79,10,10,10,13,5,,0.00,6,,110
2014-7-6,62,58,55,51,50,46,80,71,58,30.13,30.07,29.89,10,10,10,20,10,29,0.00,6,Rain,213
2014-7-7,61,57,55,56,53,51,96,87,75,30.10,30.07,30.05,10,9,4,16,4,25,0.14,8,Rain,211
2014-7-8,55,54,53,54,53,51,100,94,86,30.10,30.06,30.04,10,6,2,12,5,23,0.84,8,Rain,159
2014-7-9,57,55,53,56,54,52,100,96,83,30.24,30.18,30.11,10,7,2,9,5,,0.13,8,Rain,201
2014-7-10,61,56,53,53,52,51,100,90,75,30.23,30.17,30.03,10,8,2,8,3,,0.03,8,Rain,215
2014-7-11,57,56,54,56,54,51,100,94,84,30.02,30.00,29.98,10,5,2,12,5,,1.28,8,Rain,250
2014-7-12,59,56,55,58,56,55,100,97,93,30.18,30.06,29.99,10,6,2,15,7,26,0.32,8,Rain,275
2014-7-13,57,56,55,58,56,55,100,98,94,30.25,30.22,30.18,10,5,1,8,4,,0.29,8,Rain,291
2014-7-14,61,58,55,58,56,51,100,94,83,30.24,30.23,30.22,10,7,0,16,4,,0.01,8,Fog,307
2014-7-15,64,58,55,53,51,48,93,78,64,30.27,30.25,30.24,10,10,10,17,12,,0.00,6,,318
2014-7-16,61,56,52,51,49,47,89,76,64,30.27,30.23,30.16,10,10,10,15,6,,0.00,6,,294
2014-7-17,59,55,51,52,50,48,93,84,75,30.16,30.04,29.82,10,10,6,9,3,,0.11,7,Rain,232
2014-7-18,63,56,51,54,52,50,100,84,67,29.79,29.69,29.65,10,10,7,10,5,,0.05,6,Rain,299
2014-7-19,60,57,54,55,53,51,97,88,75,29.91,29.82,29.68,10,9,2,9,2,,0.00,8,,292
2014-7-20,57,55,52,54,52,50,94,89,77,29.92,29.87,29.78,10,8,2,13,4,,0.31,8,Rain,155
2014-7-21,69,60,52,53,51,50,97,77,52,29.99,29.88,29.78,10,10,10,13,4,,0.00,5,,297
2014-7-22,63,59,55,56,54,52,90,84,77,30.11,30.04,29.99,10,10,10,9,3,,0.00,6,Rain,240
2014-7-23,62,58,55,54,52,50,87,80,72,30.10,30.03,29.96,10,10,10,8,3,,0.00,7,,230
2014-7-24,59,57,54,54,52,51,94,84,78,29.95,29.91,29.89,10,9,3,17,4,28,0.06,8,Rain,207
2014-7-25,57,55,53,55,53,51,100,92,81,29.91,29.87,29.83,10,8,2,13,3,,0.53,8,Rain,141
2014-7-26,57,55,53,57,55,54,100,96,93,29.96,29.91,29.87,10,8,1,15,5,24,0.57,8,Rain,216
2014-7-27,61,58,55,55,54,53,100,92,78,30.10,30.05,29.97,10,9,2,13,5,,0.30,8,Rain,213
2014-7-28,59,56,53,57,54,51,97,94,90,30.06,30.00,29.96,10,8,2,9,3,,0.61,8,Rain,261
2014-7-29,61,56,51,54,52,49,96,89,75,30.13,30.02,29.95,10,9,3,14,4,,0.25,6,Rain,153
2014-7-30,61,57,54,55,53,52,97,88,78,30.31,30.23,30.14,10,10,8,8,4,,0.08,7,Rain,160
2014-7-31,66,58,50,55,52,49,100,86,65,30.31,30.29,30.26,10,9,3,10,4,,0.00,3,,217

View Code

 

highs_lows.py檔案資訊

import csv
from datetime import datetime
from matplotlib import pyplot as plt
import matplotlib as mpl

# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

# Get dates, high, and low temperatures from file.
filename = 'death_valley_2014.csv'
with open(filename) as f:
    reader = csv.reader(f)
    header_row = next(reader)
    # print(header_row)

    # for index,column_header in enumerate(header_row):
    #     print(index,column_header)

    dates, highs,lows = [],[], []
    for row in reader:
        try:
            current_date = datetime.strptime(row[0], "%Y-%m-%d")
            high = int(row[1])
            low = int(row[3])
        except ValueError: # 處理
            print(current_date, 'missing data')
        else:
            dates.append(current_date)
            highs.append(high)
            lows.append(low)

# 匯制資料圖形
fig = plt.figure(dpi=120,figsize=(10,6))
plt.plot(dates,highs,c='red',alpha=0.5)# alpha指定透明度
plt.plot(dates,lows,c='blue',alpha=0.5)
plt.fill_between(dates,highs,lows,facecolor='orange',alpha=0.1)#接收一個x值系列和y值系列，給圖表區域著色

#設定圖形格式
plt.title('2014年加利福尼亞死亡穀日氣溫最高最低圖',fontsize=24)
plt.xlabel('日（D）',fontsize=16)
fig.autofmt_xdate() # 繪製斜體日期標籤
plt.ylabel('溫度（F）',fontsize=16)
plt.tick_params(axis='both',which='major',labelsize=16)
# plt.axis([0,31,54,72]) # 自定義數軸起始刻度
plt.savefig('highs_lows.png',bbox_inches='tight')

plt.show()

6.3 生成結果

7 製作世界人口地圖：JSON格式

7.1 需求描述

下載json格式的人口資料，並使用json模組來處理。

7.2 原始碼

 json資料population_data.json部分資訊

countries.py

from pygal.maps.world import COUNTRIES

for country_code in sorted(COUNTRIES.keys()):
    print(country_code, COUNTRIES[country_code])

 countries_codes.py

from pygal.maps.world import COUNTRIES
def get_country_code(country_name):
    """Return the Pygal 2-digit country code for the given country."""
    for code, name in COUNTRIES.items():
        if name == country_name:
            return code
    # If the country wasn't found, return None.
    return

print(get_country_code('Thailand'))
# print(get_country_code('Andorra'))

americas.py

import pygal

wm =pygal.maps.world.World()
wm.title = 'North, Central, and South America'

wm.add('North America', ['ca', 'mx', 'us'])
wm.add('Central America', ['bz', 'cr', 'gt', 'hn', 'ni', 'pa', 'sv'])
wm.add('South America', ['ar', 'bo', 'br', 'cl', 'co', 'ec', 'gf',
    'gy', 'pe', 'py', 'sr', 'uy', 've'])
wm.add('Asia', ['cn', 'jp', 'th'])
wm.render_to_file('americas.svg')

world_population.py

#conding = utf-8
import json
from matplotlib import pyplot as plt
import matplotlib as mpl
from country_codes import get_country_code
import pygal
from pygal.style import RotateStyle
from pygal.style import LightColorizedStyle
# 解決中文亂碼問題
mpl.rcParams['font.sans-serif']=['SimHei']
mpl.rcParams['axes.unicode_minus']=False

# 載入json資料
filename='population_data.json'
with open(filename) as f:
    pop_data = json.load(f)
    # print(pop_data[1])


# 建立一個包含人口的字典
cc_populations={}
# cc1_populations={}

# 列印每個國家2010年的人口數量
for pop_dict in pop_data:
    if pop_dict['Year'] == '2010':
        country_name = pop_dict['Country Name']
        population = int(float(pop_dict['Value'])) # 字串數值轉化為整數
        # print(country_name + ":" + str(population))
        code = get_country_code(country_name)
        if code:
            cc_populations[code] = population
    # elif pop_dict['Year'] == '2009':
    #     country_name = pop_dict['Country Name']
    #     population = int(float(pop_dict['Value'])) # 字串數值轉化為整數
    #     # print(country_name + ":" + str(population))
    #     code = get_country_code(country_name)
    #     if code:
    #         cc1_populations[code] = population

cc_pops_1,cc_pops_2,cc_pops_3={},{},{}
for cc,pop in cc_populations.items():
    if pop <10000000:
        cc_pops_1[cc]=pop
    elif pop<1000000000:
        cc_pops_2[cc]=pop
    else:
        cc_pops_3[cc]=pop

# print(len(cc_pops_1),len(cc_pops_2),len(cc_pops_3))

wm_style = RotateStyle('#336699',base_style=LightColorizedStyle)
wm =pygal.maps.world.World(style=wm_style)
wm.title = '2010年世界各國人口統計圖'
wm.add('0-10m', cc_pops_1)
wm.add('10m-1bm',cc_pops_2)
wm.add('>1bm',cc_pops_3)
# wm.add('2009', cc1_populations)

wm.render_to_file('world_populations.svg')

7.3 生成結果

countries.py

world_population.py

8 Pygal視覺化github倉庫

8.1 需求描述

呼叫web API對GitHub資料倉庫進行視覺化展示:https://api.github.com/search/repositories?q=language:python&sort=stars

8.2 原始碼

python_repos.py

# coding=utf-8
import requests
import pygal
from pygal.style import LightColorizedStyle as LCS, LightenStyle as LS


# Make an API call, and store the response.
url = 'https://api.github.com/search/repositories?q=language:python&sort=stars'
r = requests.get(url)
print("Status code:", r.status_code) # 檢視請求是否成功，200表示成功

response_dict = r.json()
# print(response_dict.keys())
print("Total repositories:", response_dict['total_count'])

# Explore information about the repositories.
repo_dicts = response_dict['items']
print("Repositories returned:",len(repo_dicts))

# 檢視專案資訊
# repo_dict =repo_dicts[0]
# print('\n\neach repository:')
# for repo_dict in repo_dicts:
#     print("\nName:",repo_dict['name'])
#     print("Owner:",repo_dict['owner']['login'])
#     print("Stars:",repo_dict['stargazers_count'])
#     print("Repository:",repo_dict['html_url'])
#     print("Description:",repo_dict['description'])
# 檢視每個專案的鍵
# print('\nKeys:',len(repo_dict))
# for key in sorted(repo_dict.keys()):
#     print(key)

names, plot_dicts = [], []
for repo_dict in repo_dicts:
    names.append(repo_dict['name'])
    plot_dicts.append(repo_dict['stargazers_count'])

# 視覺化
my_style = LS('#333366', base_style=LCS)

my_config = pygal.Config() # Pygal類Config例項化
my_config.x_label_rotation = 45 # x軸標籤旋轉45度
my_config.show_legend = False # show_legend隱藏圖例
my_config.title_font_size = 24 # 設定圖示標題主標籤副標籤的字型大小
my_config.label_font_size = 14
my_config.major_label_font_size = 18
my_config.truncate_label = 15 # 較長的專案名稱縮短15字元
my_config.show_y_guides = False # 隱藏圖表中的水平線
my_config.width = 1000 # 自定義圖表的寬度

chart = pygal.Bar(my_config, style=my_style)
chart.title = 'Most-Starred Python Projects on GitHub'
chart.x_labels = names
chart.add('', plot_dicts)
chart.render_to_file('python_repos.svg')

8.3 生成結果

9 參考文獻