ggplot在python中的使用（plotnine）

# 条形图
median_age_dict={
    ‘Country‘: [‘New Zealand‘,‘Spain‘,‘Ireland‘,‘Israel‘,‘Denmark‘,‘Norway‘,‘Netherlands‘,‘Australia‘,‘Italy‘,‘Sweden‘],
    ‘Age‘: [39.0, 37.0, 35.0, 34.0, 34.0, 34.0, 34.0, 34.0, 34.0, 34.0]
}
median_age=pd.DataFrame(median_age_dict)

(
#     fill以颜色区分
ggplot(median_age,aes(x=‘Country‘,y=‘Age‘,fill=‘Country‘))
#     条形图，需要stat指明统计量
    +geom_bar(stat=‘identity‘,width=0.5)
#     文本标签，nudge_y表示偏离量
    +geom_text(aes(x=‘Country‘,y=‘Age‘,label=‘Age‘),nudge_y=2)
    +coord_flip()# 翻转x，y
#     排序条形图
    +xlim(median_age[‘Country‘])
#     +xlim(median_age[‘Country‘][::-1])逆序
#     隐藏图例
    +theme(legend_position = ‘none‘)
#     加上标题
    +ggtitle(‘Top 10 Median age of respondents from different countries‘)
)

# 折线图
from plotnine.data import economics
save_rate = economics[economics[‘date‘]>‘2013-01-01‘]
save_rate=save_rate.reset_index(drop=True)
(
ggplot(save_rate,aes(x=‘date‘,y=‘psavert‘))
+ geom_line(color=‘blue‘)
+geom_point(color=‘red‘)
+ ylim(0,6)#y轴的范围
#     改变x坐标刻度及间距
+ scale_x_date(breaks=‘5 months‘,date_labels=‘%Y-%m‘)
)

# 堆叠条形图
importance_dict={
    ‘tool‘: [‘Python‘,‘Python‘,‘Python‘,‘R‘,‘R‘,‘R‘,‘BigData‘,‘BigData‘,‘BigData‘,‘SQL‘,‘SQL‘,‘SQL‘],
    ‘importance‘: [‘Necessary‘,‘Nice to have‘,‘Unnecessary‘,‘Nice to have‘,‘Necessary‘,‘Unnecessary‘,‘Nice to have‘,‘Necessary‘,‘Unnecessary‘,‘Nice to have‘,‘Necessary‘,‘Unnecessary‘],
    ‘mix‘: [0.6459935499875962,0.32721409079632846,0.026792359216075416,0.5139452332657201,0.4148073022312373,0.07124746450304259,0.5740647118301314,0.3799292214357937,0.04600606673407482,0.49177800616649536,0.434224049331963,0.07399794450154162]
}
Jobskillimpotance=pd.DataFrame(importance_dict)

# Jobskillimpotance不能绘制多个折线图，因为geom_line要求数据点是数值坐标形式……
(
ggplot(Jobskillimpotance,aes(x=‘tool‘,y=‘mix‘,fill=‘importance‘))#传入数据来源和映射
+ geom_bar(stat=‘identity‘)#统计方式为原数据
)

# 多折线图
(
ggplot(df,aes(x=‘year‘,y=‘mix‘,fill=‘gender‘,color=‘gender‘))
+ geom_line(size=1)
+ geom_point(aes(shape=‘gender‘),size=3)#添加数据标记
+ scale_x_date(breaks=‘5 years‘,date_labels=‘%Y‘)#设置x轴的间隔和表示格式
+ ylim(0,6)#y轴范围
)