使用Plotly绘制基本的柱状图，需要用到的函数是graph_objs 中 Bar函数
通过参数，可以设置柱状图的样式。
通过barmod进行设置可以绘制出不同类型的柱状图出来。
我们先来实现一个简单的柱状图：
1# -*- coding: utf-8 -*- 2import plotly as py 3import plotly.graph_objs as go 4pyplt = py.offline.plot 5# Trace 6trace_basic = [go.Bar( 7 x = ['Variable_1', 'Variable_2', 'Variable_3','Variable_4','Variable_5'], 8 y = [1, 2, 3, 2, 4], 9 )] 10# Layout 11layout_basic = go.Layout( 12 title = 'The Graph Title', 13 xaxis = go.XAxis(range = [-0.5,4.5], domain = [0,1]) 14 ) 15# Figure 16figure_basic = go.Figure(data = trace_basic, layout = layout_basic) 17# Plot 18pyplt(figure_basic, filename='tmp/1.html')
上面这个例子，就是一个简单的柱状图。
下面我们讲下另外一种图，柱状簇
实现过程则是，在基本的柱状图中，加入多租数据即可实现，柱状簇
1import plotly as py 2import plotly.graph_objs as go 3pyplt = py.offline.plot 4# Traces 5trace_1 = go.Bar( 6 x = ["西南石油", "东方明珠", "海泰发展"], 7 y = [4.12, 5.32, 0.60], 8 name = "201609" 9 ) 10trace_2 = go.Bar( 11 x = ["西南石油", "东方明珠", "海泰发展"], 12 y = [3.65, 6.14, 0.58], 13 name = "201612" 14 ) 15 16trace_3 = go.Bar( 17 x = ["西南石油", "东方明珠", "海泰发展"], 18 y = [2.15, 1.35, 0.19], 19 name = "201703" 20 ) 21trace = [trace_1, trace_2, trace_3] 22# Layout 23layout = go.Layout( 24 title = '净资产收益率对比图' 25 ) 26# Figure 27figure = go.Figure(data = trace, layout = layout) 28# Plot 29pyplt(figure, filename='tmp/2.html')
执行上述代码，我们可以看到如上图所示柱状簇图例
可将数据堆叠生成。
接下来在讲讲如何绘制层叠柱状图
层叠柱状图的绘制方法与柱状簇的绘制方法基本差不多
也就是对同一个柱状簇进行叠加，实现方法是对Layout中的barmode属性进行设置
barmode = 'stack'
其余参数，与柱状簇相同。
1# -*- coding: utf-8 -*- 2import plotly as py 3import plotly.graph_objs as go 4pyplt = py.offline.plot 5 6# Stacked Bar Chart 7trace_1 = go.Bar( 8 x = ['深证50', '上证50', '西南50', '西北50','华中50'], 9 y = [0.7252, 0.9912, 0.5347, 0.4436, 0.9911], 10 name = '股票投资' 11) 12 13trace_2 = go.Bar( 14 x = ['深证50', '上证50', '西南50', '西北50','华中50'], 15 y = [0.2072, 0, 0.4081, 0.4955, 0.02], 16 name='其它投资' 17) 18 19trace_3 = go.Bar( 20 x = ['深证50', '上证50', '西南50', '西北50','华中50'], 21 y = [0, 0, 0.037, 0, 0], 22 name='债券投资' 23) 24 25trace_4 = go.Bar( 26 x = ['深证50', '上证50', '西南50', '西北50','华中50'], 27 y = [0.0676, 0.0087, 0.0202, 0.0609, 0.0087], 28 name='银行存款' 29) 30 31trace = [trace_1, trace_2, trace_3, trace_4] 32layout = go.Layout( 33 title = '基金资产配置比例图', 34 barmode='stack' 35) 36 37fig = go.Figure(data = trace, layout = layout) 38pyplt(fig, filename='tmp/1.html')
瀑布式柱状图
瀑布式柱状图是层叠柱状图的另外一种表现
可以选择性地显示层叠部分来实现柱状图的悬浮效果。
1# -*- coding: utf-8 -*- 2import plotly as py 3import plotly.graph_objs as go 4pyplt = py.offline.plot 5 6x_data = ['资产1', '资产2', 7 '资产3','资产4', '总资产'] 8y_data = [56000000, 65000000, 65000000, 81000000, 81000000] 9text = ['666,999,888万元', '8,899,666万元', '88,899,666万元', '16,167,657万元', '888,888,888万元'] 10 11# Base 12trace0 = go.Bar( 13 x=x_data, 14 y=[0, 57999848, 0, 66899764, 0], 15 marker=dict( 16 color='rgba(1,1,1, 0.0)', 17 ) 18) 19# Trace 20trace1 = go.Bar( 21 x=x_data, 22 y=[57999848, 8899916, 66899764,16167657, 83067421], 23 marker=dict( 24 color='rgba(55, 128, 191, 0.7)', 25 line=dict( 26 color='rgba(55, 128, 191, 1.0)', 27 width=2, 28 ) 29 ) 30) 31 32data = [trace0, trace1] 33layout = go.Layout( 34 title = '测试图例', 35 barmode='stack', 36 showlegend=False 37) 38 39annotations = [] 40 41for i in range(0, 5): 42 annotations.append(dict(x=x_data[i], y=y_data[i], text=text[i], 43 font=dict(family='Arial', size=14, 44 color='rgba(245, 246, 249, 1)'), 45 showarrow=False,)) 46 layout['annotations'] = annotations 47 48fig = go.Figure(data=data, layout=layout) 49pyplt(fig, filename = 'tmp/1.html')
运行上述代码，可以得到如上图所示的瀑布式柱状图。
下面我们说说，图形样式的设置。
对于柱状图颜色与样式的设置可以通过设置下面这个案例来说明。
1import plotly as py 2import plotly.graph_objs as go 3pyplt = py.offline.plot 4 5# Customizing Individual Bar Colors 6volume = [0.49,0.71,1.43,1.4,0.93] 7width = [each*3/sum(volume) for each in volume] 8trace0 = go.Bar( 9 x = ['AU.SHF', 'AG.SHF', 'SN.SHF', 10 'PB.SHF', 'CU.SHF'], 11 y = [0.85, 0.13, -0.93, 0.46, 0.06], 12 width = width, 13 marker = dict( 14 color=['rgb(205,38,38)', 'rgb(205,38,38)', 15 'rgb(34,139,34)', 'rgb(205,38,38)', 16 'rgb(205,38,38)'], 17 line=dict( 18 color='rgb(0,0,0)', 19 width=1.5, 20 )), 21 opacity = 0.8, 22) 23 24data = [trace0] 25layout = go.Layout( 26 title = '有色金属板块主力合约日内最高涨幅与波动率图', 27 xaxis=dict(tickangle=-45), 28) 29 30fig = go.Figure(data=data, layout=layout) 31pyplt(fig, filename='tmp/4.html')
运行上述代码，可以看到上图所示图例
柱状图展示了5种金属，在某个交易日的最高涨幅与波动率情况，柱形图宽度表示相对波动率的高低
柱形图越宽，波动率越大，高度表示涨幅，红色表示上涨，绿色表示下跌。
用line设置柱状图外部线框，用width设置柱状图的宽度，用opacity设置柱状图颜色的透明度情况。
基本的柱状图情况，就讲到这里。