Python数据可视化——Plotly绘制散点图、堆积柱状图、饼图、旭日图、分布图、箱线图、时间序列图、多子图、k线图

一、数据来源

整理之前的代码，发现了之前使用plotly做数据可视化的代码，贴上来做个记录。
数据集为美国不同地区各个超市的不同商品的销量数据。
需要这份数据集的可以见文末的下载地址。
本文只列举了部分图形的可视化方法，更多的可以参看Plotly官方文档

二、导入数据

import pandas as pd
import plotly as py
import plotly.graph_objs as go
import numpy as np
data=pd.read_excel('Sample - Superstore.xls')
data.head()

一共有21个特征值，具体的含义不做过多的描述，看字段名基本都能明白。

三、散点图Scatter

展示不同种类商品Category的售价Sales和利润Profit的关系

#Q1.散点图Scatter
#展示不同种类商品Category的售价Sales和利润Profit的关系
colors=['#393E46','#2BCDC2','#F66095']
labels=data['Category'].drop_duplicates(inplace=False)#获取种类数
X=[]
Y=[]
for i in labels:
    X.append(data[data['Category']==i]['Sales'])
    Y.append(data[data['Category']==i]['Profit'])
             
fig=go.Figure()
py.offline.init_notebook_mode()
for s,p,c,l in zip(X,Y,colors,labels):
    fig.add_trace(go.Scatter(
        x=s,
        y=p,
        mode='markers',
        marker_color=c,
        name=l
    ))
    
fig.update_layout(
    title='Sales & Profit of each Category',
    xaxis=dict(title='Sales'),
    yaxis=dict(title='Profit'),)

py.offline.iplot(fig)

四、堆积柱状图Stacked Bar

#Q2.堆积柱状图Stacked Bar
#展示不同商品子类的Sub-Category中不同客户类别Segment的利润总和
data2=data.groupby(['Segment','Sub-Category'])['Profit'].sum()

fig=go.Figure()

for sg,cl in zip(data2.index.levels[0],colors):
    fig.add_trace(
    go.Bar(
        x=data2[sg].index,
        y=data2[sg].values,
        name=sg,
        marker=dict(color=cl,opacity=0.65)
        )
    )
    
fig.update(layout=dict(
    title='Profit by Sub-Category & Segment',
    xaxis=dict(title='Sub-Category'),
    yaxis=dict(title='Profit'),
    barmode='stack',
))
py.offline.iplot(fig)

五、饼图

展示不同商品类的Category利润Profit占比

#Q3.饼图
#展示不同商品类的Category利润Profit占比
colors = ['gold', 'mediumturquoise', 'darkorange', 'lightgreen']
data3=data.groupby(['Category'])['Profit'].sum()

fig = go.Figure(data=[
    go.Pie(
        labels=data3.index,
        values=data3)
])

fig.update_traces(
    title='The Percentage of Profit in each Category',
    hoverinfo='label+percent+value', 
    textfont_size=15,
    marker=dict(colors=colors, line=dict(color='#000000', width=2)))
py.offline.iplot(fig)

六、旭日图Sunburst

展现美国各个地区Region各个州State的利润情况

#Q4.旭日图Sunburst
#展现美国各个地区Region各个州State的利润情况
Labels=['United States','South','West','East','Central']
Region=['South','West','East','Central']
Parents=['','United States','United States','United States','United States']
data4=data[['Region','State','Profit']]
data4_1=np.array(data4)
Values=[data4['Profit'].sum(),
        data4[data4['Region']=='South']['Profit'].sum(),
        data4[data4['Region']=='West']['Profit'].sum(),
        data4[data4['Region']=='East']['Profit'].sum(),
        data4[data4['Region']=='Central']['Profit'].sum()
       ]
for i in data4_1:
    if i[1] not in Labels:
        Labels.append(i[1])
        Parents.append(i[0])
        Values.append(data4[data4['State']==i[1]]['Profit'].sum())
        
fig=go.Figure(
    go.Sunburst(
   labels=Labels,
    parents=Parents,
    values=Values,
    ))
fig.update_layout(
    margin = dict(t=1, l=1, r=1, b=10)
)

py.offline.iplot(fig)

七、分布图Distplot

展现商品类别Category的质量Quantity分布情况

#Q5.分布图Distplot
#展现商品类别Category的质量Quantity分布情况
import plotly.figure_factory as ff

Category=data['Category'].drop_duplicates(inplace=False).tolist()

# Add histogram data
x1 =data[data['Category']==Category[0]]['Quantity']
x2 = data[data['Category']==Category[1]]['Quantity']
x3 = data[data['Category']==Category[2]]['Quantity']


# Group data together
hist_data = [x1, x2, x3]
group_labels = Category

 # Create distplot with custom bin_size
fig = ff.create_distplot(hist_data, group_labels, bin_size=[.1, .25, .5])

fig.update(layout=dict(
    title='The Quanlity of Each Category  ',
))
py.offline.iplot(fig)

八、箱线图Boxplot

展现商品类别Category的折扣Discount分布情况

#Q6.箱线图Boxplot
#展现商品类别Category的折扣Discount分布情况

fig = go.Figure()
for i in Category:
    fig.add_trace(go.Box(
        y=data[data['Category']==i]['Discount'],
        name=i))

fig.update(layout=dict(
    title='Dicount Distribution ',
))

py.offline.iplot(fig)

九、热点图Heatmap

展现商品类别Category在不同地区Region的利润Profit的相关性

#Q7.热点图Heatmap
#展现商品类别Category在不同地区Region的利润Profit的相关性

Z=[]
for i in Category:
    Y=[]
    data_0=data[data['Category']==i]
    for j in Region:
        Y.append(data_0[data_0['Region']==j]['Profit'].sum())
    Z.append(Y)
    
fig = go.Figure(data=go.Heatmap(
                   z=Z,
                   x=Region,
                   y=Category,
                   hoverongaps = False))

fig.update(layout=dict(
    title='Profit in Different Regions of Three Categories ',
))
py.offline.iplot(fig)

十、.时间序列图 Time Series(使用时间序列选择器)

展现各个年份每月的利润Profit情况。底部的时间序列选择器可以通过拖动展示指定时间区间的数据。

#Q9 .时间序列图 Time Series(使用时间序列选择器)
#展现各个年份每月的利润Profit情况
fig = go.Figure()

for value,year,color in zip(Y,Year,colors):
    fig.add_trace(go.Scatter(
                             x=Month,
                             y=value, 
                             name=year,
                             line_color=color))



fig.update_layout(
    title_text='Time Series with Rangeslider',
    xaxis=dict(title='Month'),
    yaxis=dict(title='Profit'),
    xaxis_rangeslider_visible=True)
py.offline.iplot(fig)

十一、多子图

各个商品子类Sub-Category的利润Profit

#Q10.多子图Subplots
#各个商品子类Sub-Category的利润Profit
fig = go.Figure()

trace1 = go.Bar(
        x=data2[sg].index,
        y=data.groupby(['Sub-Category'])['Sales'].sum(),
        name='Sales'
)
trace2 = go.Scatter(
    x=data2[sg].index,
    y=data.groupby(['Sub-Category'])['Profit'].sum(),
    name='Profit',
    xaxis='x', 
)
 
data = [trace1, trace2]
layout = go.Layout(
    title='Profit & Sales of each Sub-Category',
    yaxis2=dict(anchor='x', overlaying='y', side='right')#设置坐标轴的格式，一般次坐标轴在右侧
)
fig=go.Figure(data=data,layout=layout)
py.offline.iplot(fig)

十二、多图表Table and Chart Subplots

表格展现订单时间Order Date销售的物品名称Product Name以及对应的利润Profit

#Q11.多图表Table and Chart Subplots
#表格展现订单时间Order Date销售的物品名称Product Name以及对应的利润Profit
from plotly.subplots import make_subplots
data11=data[['Order Date','Product Name','Profit']]
va=data11.sort_values('Order Date')
fig = make_subplots(
    rows=2, cols=1,
    shared_xaxes=True,
    vertical_spacing=0.03,
    specs=[[{
     "type": "table"}],
           [{
     "type": "scatter"}]]
)



fig.add_trace(
    go.Scatter(
        x=va["Order Date"],
        y=va["Profit"],
        mode="lines",
        name="hash-rate-TH/s"
    ),
    row=2, col=1
)

fig.add_trace(
    go.Table(
        header=dict(
            values=['Order ID','Product Name','Profit'],
            font=dict(size=10),
            align="left"
        ),
        cells=dict(
            values=[va[k].tolist() for k in va.columns],
            align = "left")
    ),
    row=1, col=1
)
fig.update_layout(
    height=800,
    showlegend=False,
    title_text="Profits from 2015 to 2018",
    xaxis=dict(title='Year'),
    yaxis=dict(title='Profit'),
)

fig.show()

十三、k线图

十四、代码+数据集获取

关注以下公众号回复"0040"即可get√