怎么用matplotlib畫出漂亮的分析圖表

今日錦囊

特征錦囊：怎么用matplotlib畫出漂亮的分析圖表

?? Index

• 數(shù)據(jù)集引入
• 折線圖
• 餅圖
• 散點(diǎn)圖
• 面積圖
• 直方圖
• 條形圖

關(guān)于用matplotlib畫圖，先前的錦囊里有提及到，不過那些圖都是比較簡(jiǎn)陋的難登大雅之堂，作為一名優(yōu)秀的分析師，還是得學(xué)會(huì)一些讓圖表漂亮的技巧，這樣子拿出去才更加有面子哈哈。好了，今天的錦囊就是介紹一下各種常見的圖表，可以怎么來畫吧。

?? 數(shù)據(jù)集引入

首先引入數(shù)據(jù)集，我們還用一樣的數(shù)據(jù)集吧，分別是 Salary_Ranges_by_Job_Classification以及 GlobalLandTemperaturesByCity。

#?導(dǎo)入一些常用包
import?pandas?as?pd
import?numpy?as?np
import?seaborn?as?sns

%matplotlib?inline
import?matplotlib.pyplot?as?plt
import?matplotlib?as?mpl
plt.style.use('fivethirtyeight')

#解決中文顯示問題，Mac
from?matplotlib.font_manager?import?FontProperties

#?查看本機(jī)plt的有效style
print(plt.style.available)
#?根據(jù)本機(jī)available的style，選擇其中一個(gè)，因?yàn)橹爸纆gplot很好看，所以我選擇了它
mpl.style.use(['ggplot'])

#?['_classic_test',?'bmh',?'classic',?'dark_background',?'fast',?'fivethirtyeight',?'ggplot',?'grayscale',?'seaborn-bright',?'seaborn-colorblind',?'seaborn-dark-palette',?'seaborn-dark',?'seaborn-darkgrid',?'seaborn-deep',?'seaborn-muted',?'seaborn-notebook',?'seaborn-paper',?'seaborn-pastel',?'seaborn-poster',?'seaborn-talk',?'seaborn-ticks',?'seaborn-white',?'seaborn-whitegrid',?'seaborn',?'Solarize_Light2']

#?數(shù)據(jù)集導(dǎo)入

#?引入第?1?個(gè)數(shù)據(jù)集?Salary_Ranges_by_Job_Classification
salary_ranges?=?pd.read_csv('./data/Salary_Ranges_by_Job_Classification.csv')

#?引入第?2?個(gè)數(shù)據(jù)集?GlobalLandTemperaturesByCity
climate?=?pd.read_csv('./data/GlobalLandTemperaturesByCity.csv')
#?移除缺失值
climate.dropna(axis=0,?inplace=True)
#?只看中國(guó)
#?日期轉(zhuǎn)換,?將dt?轉(zhuǎn)換為日期，取年份,?注意map的用法
climate['dt']?=?pd.to_datetime(climate['dt'])
climate['year']?=?climate['dt'].map(lambda?value:?value.year)
climate_sub_china?=?climate.loc[climate['Country']?==?'China']
climate_sub_china['Century']?=?climate_sub_china['year'].map(lambda?x:int(x/100?+1))
climate.head()

?? 折線圖

折線圖是比較簡(jiǎn)單的圖表了，也沒有什么好優(yōu)化的，顏色看起來順眼就好了。下面是從網(wǎng)上找到了顏色表，可以從中挑選～

#?選擇上海部分天氣數(shù)據(jù)
df1?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shanghai')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.set_index('dt')
df1.head()

#?折線圖
df1.plot(colors=['lime'])
plt.title('AverageTemperature?Of?ShangHai')
plt.ylabel('Number?of?immigrants')
plt.xlabel('Years')
plt.show()

上面這是單條折線圖，多條折線圖也是可以畫的，只需要多增加幾列。

#?多條折線圖
df1?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shanghai')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'SH'})
df2?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Tianjin')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'TJ'})
df3?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shenyang')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'SY'})
#?合并
df123?=?df1.merge(df2,?how='inner',?on=['dt'])\
????????????????.merge(df3,?how='inner',?on=['dt'])\
????????????????.set_index(['dt'])
df123.head()

#?多條折線圖
df123.plot()
plt.title('AverageTemperature?Of?3?City')
plt.ylabel('Number?of?immigrants')
plt.xlabel('Years')
plt.show()

?? 餅圖

接下來是畫餅圖，我們可以優(yōu)化的點(diǎn)多了一些，比如說從餅塊的分離程度，我們先畫一個(gè)“低配版”的餅圖。

df1?=?salary_ranges.groupby('SetID',?axis=0).sum()

#?“低配版”餅圖
df1['Step'].plot(kind='pie',?figsize=(7,7),
??????????????????autopct='%1.1f%%',
??????????????????shadow=True)
plt.axis('equal')
plt.show()

#?“高配版”餅圖
colors?=?['lightgreen',?'lightblue']?#控制餅圖顏色?['lightgreen',?'lightblue',?'pink',?'purple',?'grey',?'gold']
explode=[0,?0.2]?#控制餅圖分離狀態(tài)，越大越分離

df1['Step'].plot(kind='pie',?figsize=(7,?7),
??????????????????autopct?=?'%1.1f%%',?startangle=90,
??????????????????shadow=True,?labels=None,?pctdistance=1.12,?colors=colors,?explode?=?explode)
plt.axis('equal')
plt.legend(labels=df1.index,?loc='upper?right',?fontsize=14)
plt.show()

?? 散點(diǎn)圖

散點(diǎn)圖可以優(yōu)化的地方比較少了，ggplot2的配色都蠻好看的，正所謂style選的好，省很多功夫！

#?選擇上海部分天氣數(shù)據(jù)
df1?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shanghai')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'SH'})

df2?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shenyang')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'SY'})
#?合并
df12?=?df1.merge(df2,?how='inner',?on=['dt'])
df12.head()

#?散點(diǎn)圖
df12.plot(kind='scatter',??x='SH',?y='SY',?figsize=(10,?6),?color='darkred')
plt.title('Average?Temperature?Between?ShangHai?-?ShenYang')
plt.xlabel('ShangHai')
plt.ylabel('ShenYang')
plt.show()

?? 面積圖

#?多條折線圖
df1?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shanghai')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'SH'})
df2?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Tianjin')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'TJ'})
df3?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shenyang')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.rename(columns={'AverageTemperature':'SY'})
#?合并
df123?=?df1.merge(df2,?how='inner',?on=['dt'])\
????????????????.merge(df3,?how='inner',?on=['dt'])\
????????????????.set_index(['dt'])
df123.head()

colors?=?['red',?'pink',?'blue']?#控制餅圖顏色?['lightgreen',?'lightblue',?'pink',?'purple',?'grey',?'gold']
df123.plot(kind='area',?stacked=False,
????????figsize=(20,?10),?colors=colors)
plt.title('AverageTemperature?Of?3?City')
plt.ylabel('AverageTemperature')
plt.xlabel('Years')
plt.show()

?? 直方圖

#?選擇上海部分天氣數(shù)據(jù)
df?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shanghai')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.set_index('dt')
df.head()

#?最簡(jiǎn)單的直方圖
df['AverageTemperature'].plot(kind='hist',?figsize=(8,5),?colors=['grey'])
plt.title('ShangHai?AverageTemperature?Of?2010-2013')?#?add?a?title?to?the?histogram
plt.ylabel('Number?of?month')?#?add?y-label
plt.xlabel('AverageTemperature')?#?add?x-label
plt.show()

?? 條形圖

#?選擇上海部分天氣數(shù)據(jù)
df?=?climate.loc[(climate['Country']=='China')&(climate['City']=='Shanghai')&(climate['dt']>='2010-01-01')]\
??????????????????.loc[:,['dt','AverageTemperature']]\
??????????????????.set_index('dt')
df.head()

df.plot(kind='bar',?figsize?=?(10,?6))
plt.xlabel('Month')?
plt.ylabel('AverageTemperature')?
plt.title('AverageTemperature?of?shanghai')
plt.show()

df.plot(kind='barh',?figsize=(12,?16),?color='steelblue')
plt.xlabel('AverageTemperature')?
plt.ylabel('Month')?
plt.title('AverageTemperature?of?shanghai')?
plt.show()

今天的內(nèi)容比較長(zhǎng)了，建議收藏起來哦，下次有空的時(shí)候可以把它弄進(jìn)自己的代碼庫(kù)，使用起來更加方便哦～