干货分享 | 用Pyecharts制作炫酷的可视化大屏
首先介绍Pyecharts模块当中的Grid组件，使用Grid组件可以很好地将多张图无论是上下组合还是左右组合，都能够很好地拼接起来，我们先来看第一个例子
bar = (Bar().add_xaxis(Faker.choose()).add_yaxis("商家1", Faker.values()).add_yaxis("商家2", Faker.values()).set_global_opts(title_opts=opts.TitleOpts(title="直方图")))line = (Line().add_xaxis(Faker.choose()).add_yaxis("商家1", Faker.values()).add_yaxis("商家2", Faker.values()).set_global_opts(title_opts=opts.TitleOpts(title="折线图", pos_top="48%"),legend_opts=opts.LegendOpts(pos_top="48%"),))grid = (Grid().add(bar, grid_opts=opts.GridOpts(pos_bottom="60%")).add(line, grid_opts=opts.GridOpts(pos_top="50%")).render("水平组合图_test.html"))
我们可以看到两张图表被以上下组合的方式拼接起来，当然除了上下的拼接以外，我们还可以左右来拼接，代码如下
bar = (Bar().add_xaxis(Faker.choose()).add_yaxis("商家1", Faker.values()).add_yaxis("商家2", Faker.values()).set_global_opts(title_opts=opts.TitleOpts(title="直方图"),legend_opts=opts.LegendOpts(pos_left="20%"),))line = (Line().add_xaxis(Faker.choose()).add_yaxis("商家1", Faker.values()).add_yaxis("商家2", Faker.values()).set_global_opts(title_opts=opts.TitleOpts(title="折线图", pos_right="5%"),legend_opts=opts.LegendOpts(pos_right="20%"),))grid = (Grid().add(bar, grid_opts=opts.GridOpts(pos_left="60%")).add(line, grid_opts=opts.GridOpts(pos_right="50%")).render("垂直组合图_test.html"))
可以看到我们无论是想上下拼接还是左右拼接，都可以通过调整参数“pos_left”、“pos_right”、“pos_top”以及“pos_bottom”这几个参数来实现，我们再来看一下下面这个例子，我们也可以将地图和直方图两者拼接起来
bar = (Bar().add_xaxis(Faker.choose()).add_yaxis("商家1", Faker.values()).add_yaxis("商家2", Faker.values()).set_global_opts(legend_opts=opts.LegendOpts(pos_left="20%")))map = (Map().add("商家1", [list(z) for z in zip(Faker.provinces, Faker.values())], "china").set_global_opts(title_opts=opts.TitleOpts(title="地图-基本示例")))grid = (Grid().add(bar, grid_opts=opts.GridOpts(pos_top="50%", pos_right="75%")).add(map, grid_opts=opts.GridOpts(pos_left="60%")).render("地图+直方图.html"))
英文单词“overlap”的意思是重叠，那么放在这里，也就指的是可以将多张图合并成一张，那么该怎么结合才好呢？我们来看一下下面这个例子，我们将直方图和折线图通过overlap组件组合到一起
v1 = Faker.values()v2 = Faker.values()v3 = Faker.values()bar = (Bar().add_xaxis(Faker.provinces).add_yaxis("商家A", v1).add_yaxis("商家B", v2).extend_axis(yaxis=opts.AxisOpts(axislabel_opts=opts.LabelOpts(formatter="{value} 个"), interval=20)).set_series_opts(label_opts=opts.LabelOpts(is_show=False)).set_global_opts(title_opts=opts.TitleOpts(title="Overlap-bar+line"),yaxis_opts=opts.AxisOpts(axislabel_opts=opts.LabelOpts(formatter="{value} 个")),))line = Line().add_xaxis(Faker.provinces).add_yaxis("商家C", v3, yaxis_index=1)bar.overlap(line)bar.render("直方图+折线图Overlap.html")
除此之外，我们也可以将散点图和折线图合并在一张图上面，在代码上就只要将直方图的代码替换成散点图的就行，这边也就具体不做演示
Bar().add_xaxis(x_data).add_yaxis("A",[],yaxis_index=0,color="#d14a61",).add_yaxis("B",[],yaxis_index=1,color="#5793f3",).直方图的全局配置代码....line = (Line().add_xaxis(x_data).add_yaxis("C",[],yaxis_index=2,color="#675bba",label_opts=opts.LabelOpts(is_show=False),))bar.overlap(line)grid = Grid()grid.add(bar, opts.GridOpts(pos_left="5%", pos_right="20%"), is_control_axis_index=True)grid.render("test.html")
我们在用Pyecharts绘制了多张图表之后，可以直接Tab组件将多张图表连起来，一页放一张图表，具体看下面的例子和代码，
def bar_datazoom_slider() -> Bar:c = (Bar().add_xaxis(Faker.days_attrs).add_yaxis("商家A", Faker.days_values).set_global_opts(title_opts=opts.TitleOpts(title="Bar-DataZoom（slider-水平）"),datazoom_opts=[opts.DataZoomOpts()],))return cdef line_markpoint() -> Line:c = (Line().add_xaxis(Faker.choose()).add_yaxis("商家A",Faker.values(),markpoint_opts=opts.MarkPointOpts(data=[opts.MarkPointItem(type_="min")]),).set_global_opts(title_opts=opts.TitleOpts(title="折线图")))return cdef pie_rosetype() -> Pie:v = Faker.choose()c = (Pie().add("",[list(z) for z in zip(v, Faker.values())],radius=["30%", "75%"],center=["25%", "50%"],rosetype="radius",label_opts=opts.LabelOpts(is_show=False),).set_global_opts(title_opts=opts.TitleOpts(title="饼图-玫瑰图示例")))return ctab = Tab()tab.add(bar_datazoom_slider(), "直方图")tab.add(line_markpoint(), "折线图")tab.add(pie_rosetype(), "饼图")tab.render("tab_base.html")
和上面Tab组件不一样的是，Tab组件是一页放一张图表，有几张图表就分成几页，而Page组件则是将绘制完成的多张图表统统放在一张页面里面，代码的改动上面也十分的简单，只要将上面代码的Tab部分改成Page()即可，如下
def bar_datazoom_slider() -> Bar:c = (Bar().add_xaxis(Faker.days_attrs).add_yaxis("商家A", Faker.days_values).set_global_opts(title_opts=opts.TitleOpts(title="Bar-DataZoom（slider-水平）"),datazoom_opts=[opts.DataZoomOpts()],))return cdef line_markpoint() -> Line:c = (Line().add_xaxis(Faker.choose()).add_yaxis("商家A",Faker.values(),markpoint_opts=opts.MarkPointOpts(data=[opts.MarkPointItem(type_="min")]),).set_global_opts(title_opts=opts.TitleOpts(title="折线图")))return cdef pie_rosetype() -> Pie:v = Faker.choose()c = (Pie().add("",[list(z) for z in zip(v, Faker.values())],radius=["30%", "75%"],center=["25%", "50%"],rosetype="radius",label_opts=opts.LabelOpts(is_show=False),).set_global_opts(title_opts=opts.TitleOpts(title="饼图-玫瑰图示例")))return cpage = Page(layout=Page.SimplePageLayout)page.add(bar_datazoom_slider(),line_markpoint(),pie_rosetype(),)page.render("page_simple_layout.html")
上图所示的图表在页面当中是不能被挪动的，在Page()组件当中我们还能够使得图表按照我们所想的那样随意的挪动
## 上面的代码都一样，page = Page(layout=Page.DraggablePageLayout)page.add(bar_datazoom_slider(),line_markpoint(),pie_rosetype(),)page.render("page_draggable_layout.html")
评论