基本图形

柱状图

reviews['points'].value_counts().sort_index().plot.bar()

散点图

reviews[reviews['price'] < 100].sample(100).plot.scatter(x='price', y='points')

image

蜂窝图

reviews[reviews['price'] < 100].plot.hexbin(x='price', y='points', gridsize=15)

image

大量重复的点可以用这种图表示

柱状图-叠加模式

image

wine_counts.plot.bar(stacked=True)

image

面积模式

wine_counts.plot.area()

折线模式

wine_counts.plot.line()

美化

设置图的大小，字体大小，颜色，标题

reviews['points'].value_counts().sort_index().plot.bar(
    figsize=(12, 6),
    color='mediumvioletred',
    fontsize=16,
    title='Rankings Given by Wine Magazine',
)

借助Matplotlib

import matplotlib.pyplot as plt

ax = reviews['points'].value_counts().sort_index().plot.bar(
    figsize=(12, 6),
    color='mediumvioletred',
    fontsize=16
)
ax.set_title("Rankings Given by Wine Magazine", fontsize=20)

image

借助Seaborn-去除边框

import matplotlib.pyplot as plt
import seaborn as sns

ax = reviews['points'].value_counts().sort_index().plot.bar(
    figsize=(12, 6),
    color='mediumvioletred',
    fontsize=16
)
ax.set_title("Rankings Given by Wine Magazine", fontsize=20)
sns.despine(bottom=True, left=True)

image

多图表

matplotlib

fig, axarr = plt.subplots(2, 2, figsize=(12, 8))

reviews['points'].value_counts().sort_index().plot.bar(
    ax=axarr[0][0]
)

reviews['province'].value_counts().head(20).plot.bar(
    ax=axarr[1][1]

image

Seaborn

df = footballers[footballers['Position'].isin(['ST', 'GK'])]
g = sns.FacetGrid(df, col="Position", col_wrap=2)
g.map(sns.kdeplot, "Overall")

image

df = footballers[footballers['Position'].isin(['ST', 'GK'])]
df = df[df['Club'].isin(['Real Madrid CF', 'FC Barcelona', 'Atlético Madrid'])]

g = sns.FacetGrid(df, row="Position", col="Club")
g.map(sns.violinplot, "Overall")

image

df = footballers[footballers['Position'].isin(['ST', 'GK'])]
df = df[df['Club'].isin(['Real Madrid CF', 'FC Barcelona', 'Atlético Madrid'])]

g = sns.FacetGrid(df, row="Position", col="Club", 
                  row_order=['GK', 'ST'],
                  col_order=['Atlético Madrid', 'FC Barcelona', 'Real Madrid CF'])
g.map(sns.violinplot, "Overall")

控制显示顺序

pairplot-多变量的相互关系

sns.pairplot(footballers[['Overall', 'Potential', 'Value']])

image

颜色，图标参数

sns.lmplot(
  x='Value', y='Overall', 
  markers=['o', 'x', '*'], 
  hue='Position', 
  data=footballers.loc[footballers['Position'].isin(
    ['ST', 'RW', 'LW'])],
  fit_reg=False
)

image

分组

f = (footballers
         .loc[footballers['Position'].isin(['ST', 'GK'])]
         .loc[:, ['Value', 'Overall', 'Aggression', 'Position']]
    )
f = f[f["Overall"] >= 80]
f = f[f["Overall"] < 85]
f['Aggression'] = f['Aggression'].astype(float)

sns.boxplot(x="Overall", y="Aggression", hue='Position', data=f)

image

总结图

热力图

f = (
    footballers.loc[:, ['Acceleration', 'Aggression', 'Agility', 'Balance', 'Ball control']]
        .applymap(lambda v: int(v) if str.isdecimal(v) else np.nan)
        .dropna()
).corr()

sns.heatmap(f, annot=True)

image

平行线图

from pandas.plotting import parallel_coordinates

f = (
    footballers.iloc[:, 12:17]
        .loc[footballers['Position'].isin(['ST', 'GK'])]
        .applymap(lambda v: int(v) if str.isdecimal(v) else np.nan)
        .dropna()
)
f['Position'] = footballers['Position']
f = f.sample(200)

parallel_coordinates(f, 'Position'）

image

Seanborn使用

基本图形

柱状图-值统计

countplot == value_count

sns.countplot(reviews['points'])

image

折线图-密度图

sns.kdeplot(reviews.query('price < 200').price)

image

二维密度图--类似蜂窝图作用

样本多，重复点多的时候用

sns.kdeplot(reviews[reviews['price'] < 200].loc[:, ['price', 'points']].dropna().sample(5000))

image

直方图

类似pandas.hist

sns.distplot(reviews['points'], bins=10, kde=False)

image

散点图和直方图复合

sns.jointplot(x='price', y='points', data=reviews[reviews['price'] < 100])

image

蜂窝图和直方图复合

sns.jointplot(x='price', y='points', data=reviews[reviews['price'] < 100], kind='hex',gridsize=20)

image

箱线图

df = reviews[reviews.variety.isin(reviews.variety.value_counts().head(5).index)]
sns.boxplot(
    x='variety',
    y='points',
    data=df
)

image

小提琴图

sns.violinplot(
    x='variety',
    y='points',
    data=reviews[reviews.variety.isin(reviews.variety.value_counts()[:5].index)]
)

image

网络动态图表-plotly

from plotly.offline import init_notebook_mode, iplot
init_notebook_mode(connected=True)

散点图

import plotly.graph_objs as go

iplot([go.Scatter(x=reviews.head(1000)['points'], y=reviews.head(1000)['price'], mode='markers')])

image

热力图

iplot([go.Histogram2dContour(x=reviews.head(500)['points'], 
                             y=reviews.head(500)['price'], 
                             contours=go.Contours(coloring='heatmap')),
       go.Scatter(x=reviews.head(1000)['points'], y=reviews.head(1000)['price'], mode='markers')])

image

图形语法的可视化库plotnine

from plotnine import *

top_wines = reviews[reviews['variety'].isin(reviews['variety'].value_counts().head(5).index)]

df = top_wines.head(1000).dropna()

(ggplot(df)
 + aes('points', 'price')
 + geom_point())

#其他表达形式ggplot(df)
 + geom_point(aes('points', 'price'))
)

(ggplot(df, aes('points', 'price'))
 + geom_point

一层层添加图形参数

image

df = top_wines.head(1000).dropna()

(
    ggplot(df)
        + aes('points', 'price')
        + geom_point()
        + stat_smooth()
)

image

添加颜色

df = top_wines.head(1000).dropna()

(
    ggplot(df)
        + geom_point()
        + aes(color='points')
        + aes('points', 'price')
        + stat_smooth()
)

一图多表

df = top_wines.head(1000).dropna()

(ggplot(df)
     + aes('points', 'price')
     + aes(color='points')
     + geom_point()
     + stat_smooth()
     + facet_wrap('~variety')
)

image

柱状图

(ggplot(top_wines)
     + aes('points')
     + geom_bar()
)

image

二维热力图

(ggplot(top_wines)
     + aes('points', 'variety')
     + geom_bin2d(bins=20)
)

image

更多API文档 API Reference.

处理时间序列

一般柱状图

shelter_outcomes['date_of_birth'].value_counts().sort_values().plot.line()

image

按年份重新取样

shelter_outcomes['date_of_birth'].value_counts().resample('Y').sum().plot.line()

image

stocks['volume'].resample('Y').mean().plot.bar()

image

同期对比

如今年12月和去年12月比较

from pandas.plotting import lag_plot

lag_plot(stocks['volume'].tail(250))

image

自相关图

from pandas.plotting import autocorrelation_plot

autocorrelation_plot(stocks['volume'])

image