ggplot在python中的使用（plotnine）
阿新 • • 發佈：2020-09-06
# 條形圖
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以顏色區分
ggplot(median_age,aes(x='Country',y='Age',fill='Country'))
# 條形圖，需要stat指明統計量
+geom_bar(stat='identity',width=0.5)
# 文字標籤，nudge_y表示偏離量
+geom_text(aes(x='Country',y='Age',label='Age'),nudge_y=2)
+coord_flip()# 翻轉x，y
# 排序條形圖
+xlim(median_age['Country'])
# +xlim(median_age['Country'][::-1])逆序
# 隱藏圖例
+theme(legend_position = 'none')
# 加上標題
+ggtitle('Top 10 Median age of respondents from different countries')
)
# 折線圖
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軸的範圍
# 改變x座標刻度及間距
+ scale_x_date(breaks='5 months',date_labels='%Y-%m')
)
# 堆疊條形圖
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不能繪製多個折線圖，因為geom_line要求資料點是數值座標形式……
(
ggplot(Jobskillimpotance,aes(x='tool',y='mix',fill='importance'))#傳入資料來源和對映
+ geom_bar(stat='identity')#統計方式為原資料
)
# 多折線圖
(
ggplot(df,aes(x='year',y='mix',fill='gender',color='gender'))
+ geom_line(size=1)
+ geom_point(aes(shape='gender'),size=3)#新增資料標記
+ scale_x_date(breaks='5 years',date_labels='%Y')#設定x軸的間隔和表示格式
+ ylim(0,6)#y軸範圍
)