【Dash系列】Python的Web可视化框架Dash(1)---简介
【Dash系列】Python的Web可视化框架Dash(2)---安装
【Dash系列】Python的Web可视化框架Dash(3)---布局设置
【Dash系列】Python的Web可视化框架Dash(4)---基本回调
【Dash系列】Python的Web可视化框架Dash(5)---状态和预更新
【Dash系列】Python的Web可视化框架Dash(6)---交互和过滤
【Dash系列】Python的Web可视化框架Dash(7)---回调共享
【Dash系列】Python的Web可视化框架Dash(8)---核心组件
本节介绍如何实现Dash应用的交互,先导入本节用到的所有包
import arrow as ar
import numpy as np
import pandas as pd
import json
from textwrap import dedent as d
import plotly.graph_objs as go
import dash
import dash_core_components as dcc # 交互式组件
import dash_html_components as html # 代码转html
from dash.dependencies import Input, Output, State # 回调
from jupyter_plotly_dash import JupyterDash # Jupyter中的Dash
import ipywidgets as widgets # Jupyter的滑动条插件
一、概述
dash_core_components 库包含一个叫
Graph
的组件;Graph 组件使用开源的 plotly.js (JavaScript图形库) 渲染交互式数据的可视化;
Plotly.js 支持超过35种数据图,可以生成高清的SVG矢量图和高性能的WebGL图;
dcc.Graph 组件的
figure
与 Plotly.py的figure
使用一样的参数。详情可参阅plotly.py文档与图库;Dash 组件由一组属性以声明方式进行描述,所有这些属性都可以通过回调函数进行更新。有些属性还可以通过用户交互进行更新,比如,点选下拉按钮 dcc.Dropdown 组件的选项,该组件的属性 value 值就会改变;
dcc.Graph 组件有四个属性,可以通过用户交互进行更新:鼠标悬停(hoverData)、单击(clickData)、选择区域数据(selectedData)、更改布局(relayoutData)。
二、打印属性
(一) 代码
app = JupyterDash('Print_id', height=1000, width = 1000)
styles = dict(pre = dict(border = 'thin lightgrey solid', overflowX = 'scroll'))
app.layout = html.Div([
dcc.Graph(
id = 'basic-interactions',
figure = dict(data = [dict(x = [1, 2, 3, 4],
y = [4, 1, 3, 5],
text = ['a', 'b', 'c', 'd'],
customdata = ['c.a', 'c.b', 'c.c', 'c.d'],
name = 'trace01',
mode = 'markers',
marker = dict(size = 12)),
dict(x = [1, 2, 3, 4],
y = [9, 4, 1, 4],
text = ['w', 'x', 'y', 'z'],
customdata = ['c.w', 'c.x', 'c.y', 'c.z'],
name = 'trace02',
mode = 'markers',
marker = dict(size = 12))],
layout = dict(clickmode = 'event+select'))),
html.Div(className = 'row',
children = [
html.Div(className = 'three columns',
children = [
dcc.Markdown(d("""
**一、悬停数据**
将鼠标悬停在图中的值上。
""")),
html.Pre(id = 'hover-data', style = styles['pre'])]),
html.Div(className = 'three columns',
children = [
dcc.Markdown(d("""
**二、点击数据**
用鼠标点击图上的点。
""")),
html.Pre(id = 'click-data', style = styles['pre'])]),
html.Div(className = 'three columns',
children = [
dcc.Markdown(d("""
**三、选择数据**
使用菜单的套索或方框工具,选择图上的点。
请注意:如果layout.clickmode ='event + select',
若单击的同时按住shift键,则选择数据的同时,也会累计或取消累计所选的数据。
""")),
html.Pre(id = 'selected-data', style = styles['pre'])]),
html.Div(className = 'three columns',
children = [
dcc.Markdown(d("""
**四、缩放与改变数据布局**
在图形上点击并拖拽,
或点击图形菜单的缩放按钮实现缩放。
点击图例也可以激活此事件。
""")),
html.Pre(id = 'relayout-data', style = styles['pre'])])
]
)
])
@app.callback(Output('hover-data', 'children'), [Input('basic-interactions', 'hoverData')])
def display_hover_data(hoverData):
return json.dumps(hoverData, indent = 2)
@app.callback(Output('click-data', 'children'), [Input('basic-interactions', 'clickData')])
def display_click_data(clickData):
return json.dumps(clickData, indent = 2)
@app.callback(Output('selected-data', 'children'), [Input('basic-interactions', 'selectedData')])
def display_selected_data(selectedData):
return json.dumps(selectedData, indent = 2)
@app.callback(Output('relayout-data', 'children'), [Input('basic-interactions', 'relayoutData')])
def display_repayout_data(relayoutData):
return json.dumps(relayoutData, indent = 2)
app
(二) 示意图
三、鼠标悬停时更新图表
(一) 代码
df = pd.read_csv(
'https://gist.githubusercontent.com/chriddyp/'
'cb5392c35661370d95f300086accea51/raw/'
'8e0768211f6b747c0db42a9ce9a0937dafcbd8b2/'
'indicators.csv')
app = JupyterDash('update_graph_id', width = 1000, height = 960)
app.layout = html.Div([
# 设置2个下拉按钮、4个单选按钮
html.Div([
html.Div([
dcc.Dropdown(
id = 'crossfilter-xaxis-column',
options = [{'label': i, 'value': i} for i in df['Indicator Name'].unique()],
value = 'Fertility rate, total (births per woman)'),
dcc.RadioItems(
id = 'crossfilter-xaxis-type',
options = [{'label': i, 'value': i} for i in ['线性', '日志']],
value = '线性',
labelStyle = dict(display = 'inline-block'))],
style = dict(width = '49%', display = 'inline-block')),
html.Div([
dcc.Dropdown(
id = 'crossfilter-yaxis-column',
options = [{'label': i, 'value': i} for i in df['Indicator Name'].unique()],
value = 'Life expectancy at birth, total (years)'),
dcc.RadioItems(
id = 'crossfilter-yaxis-type',
options = [{'label': i, 'value': i} for i in ['线性', '日志']],
value = '线性',
labelStyle = dict(display = 'inline-block'))],
style = dict(width = '49%', display = 'inline-block', float = 'right'))],
style = dict(borderBottom = 'thin lightgrey solid',
backgroundColor = 'rgb(250, 250, 250)', padding = '10px 5px')),
# 设置交互数据对象及默认值
html.Div([
dcc.Graph(
id = 'crossfilter-indicator-scatter',
hoverData = dict(points = [{'customdata': 'Japan'}]))],
style = dict(width = '49%', display = 'inline-block', padding = '0 20')),
# 设置交互的子图表
html.Div([
dcc.Graph(id = 'x-time-series'),
dcc.Graph(id = 'y-time-series')],
style = dict(width = '49%', display = 'inline-block')),
# 设置日期滑动条
html.Div(
dcc.Slider(
id = 'crossfilter-year--slider',
min = df['Year'].min(),
max = df['Year'].max(),
value = df['Year'].max(),
marks = {str(y): str(y) for y in df['Year'].unique()}),
style = dict(widht = '49%', padding = '0px 20px 20px 20px')
)
])
# 回调2个下拉按钮 + 4个单选按钮
@app.callback(
Output('crossfilter-indicator-scatter', 'figure'),
[Input('crossfilter-xaxis-column', 'value'),
Input('crossfilter-yaxis-column', 'value'),
Input('crossfilter-xaxis-type', 'value'),
Input('crossfilter-yaxis-type', 'value'),
Input('crossfilter-year--slider', 'value')])
def update_graph(xaxis_column_name, yaxis_column_name, xaxis_type, yaxis_type, year_value):
dff = df[df['Year'] == year_value]
return dict(
data = [go.Scatter(
x = dff[dff['Indicator Name'] == xaxis_column_name]['Value'],
y = dff[dff['Indicator Name'] == yaxis_column_name]['Value'],
text = dff[dff['Indicator Name'] == yaxis_column_name]['Country Name'],
customdata = dff[dff['Indicator Name'] == yaxis_column_name]['Country Name'],
mode = 'markers',
marker = dict(size = 15, opacity = 0.5, line = dict(width = 0.5, color = 'white')))],
layout = dict(
xaxis = dict(title = xaxis_column_name, type = 'linear' if xaxis_type == '线性' else '日志'),
yaxis = dict(title = yaxis_column_name, type = 'linear' if yaxis_type == '线性' else '日志'),
margin = {'l': 40, 'b': 30, 't': 10, 'r': 0}, height=720, hovermode = 'closest'
)
)
# 设置子图表
def create_time_series(df, axis_type, title):
return dict(
data = [go.Scatter(
x = df['Year'],
y = df['Value'],
mode = 'lines+markers')],
layout = dict(
height = 360, margin = {'l': 20, 'b': 30, 't': 10, 'r': 10},
xaxis = dict(showgrid = False),
yaxis = dict(type = 'linear' if axis_type == '线性' else '日志'),
annotations = [
dict(x = 0, y = 0.85, text = title,
showarrow = False, align = 'left',
bgcolor = 'rgba(255, 255, 255, 0.5)',
xref = 'paper', yref = 'paper',
xanchor = 'left', yanchor = 'bottom')]
)
)
# 回调--设置上子图表的交互
@app.callback(
Output('x-time-series', 'figure'),
[Input('crossfilter-indicator-scatter', 'hoverData'),
Input('crossfilter-xaxis-column', 'value'),
Input('crossfilter-xaxis-type', 'value')])
def update_y_timeseries(hoverData, xaxis_column_name, axis_type):
country_name = hoverData['points'][0]['customdata']
dff = df[df['Country Name'] == country_name]
dff = dff[dff['Indicator Name'] == xaxis_column_name]
title = f"<b>{country_name}</b><br>{xaxis_column_name}"
return create_time_series(dff, axis_type, title)
# 回调--设置下子图表的交互
@app.callback(
Output('y-time-series', 'figure'),
[Input('crossfilter-indicator-scatter', 'hoverData'),
Input('crossfilter-yaxis-column', 'value'),
Input('crossfilter-yaxis-type', 'value')])
def update_x_timeseries(hoverData, yaxis_column_name, axis_type):
dff = df[df['Country Name'] == hoverData['points'][0]['customdata']]
dff = dff[dff['Indicator Name'] == yaxis_column_name]
return create_time_series(dff, axis_type, yaxis_column_name)
app
(二) 示意图
- 在左侧图表区域滑动鼠标,会更新右侧的图表
-
效果
四、通用的交叉筛选器
(一) 代码
# 数据源
np.random.seed(0) # 使每次生成的随机数相同
df = pd.DataFrame({f"Column {i}": np.random.rand(30) + i*10 for i in range(6)})
app = JupyterDash('Cross_filter_id', width = 1000)
app.layout = html.Div(
className = 'row',
children = [
html.Div(dcc.Graph(id = 'g1', config = dict(displayModeBar = False)), className = 'four columns'),
html.Div(dcc.Graph(id = 'g2', config = dict(displayModeBar = False)), className = 'four columns'),
html.Div(dcc.Graph(id = 'g3', config = dict(displayModeBar = False)), className = 'four columns'),
]
)
# high_light为返回函数的函数
def high_light(x, y):
def callback(*selectedDatas):
selectedpoints = df.index
# 按选定的点,从Dataframe中过滤数据
for i, selected_data in enumerate(selectedDatas):
if selected_data is not None:
selected_index = [p['customdata'] for p in selected_data['points']]
if len(selected_index) > 0:
selectedpoints = np.intersect1d(selectedpoints, selected_index)
# 设置图表
fig = dict(
data = [dict(
x = df[x], y = df[y], text = df.index, textposition = 'top', selectedpoints = selectedpoints,
customdata = df.index, type = 'scatter', mode = 'markers+text',
textfont = dict(color = 'rgba(30, 30, 30, 1)'),
marker = dict(color = 'rgba(0, 116, 217, 0.7)', size = 12,
line = dict(color = 'rgb(0, 116, 217)', width = 0.5)),
unselected = dict(marker = dict(opacity = 0.3), textfont = dict(color = 'rgba(0, 0, 0, 0)'))
)], # 使未选择的透明
layout = dict(clickmode = 'event+select', showlegend = False, hovermode = 'closest', height = 230,
margin = {'l': 15, 'r': 0, 'b': 15, 't': 5}, dragmode = 'select')
)
# 将矩形显示到先前选定的区域
shape = dict(type = 'rect', line = dict(width = 1, dash = 'dot', color = 'darkgrey'))
if selectedDatas[0] and selectedDatas[0]['range']:
fig['layout']['shapes'] = [dict({
'x0': selectedDatas[0]['range']['x'][0],
'x1': selectedDatas[0]['range']['x'][1],
'y0': selectedDatas[0]['range']['y'][0],
'y1': selectedDatas[0]['range']['y'][1]},
**shape)]
else:
fig['layout']['shapes'] = [dict({
'type': 'rect',
'x0': np.min(df[x]),
'x1': np.max(df[x]),
'y0': np.min(df[y]),
'y1': np.max(df[y])},
**shape)]
return fig
return callback
# app.callback是一个装饰器
app.callback(
Output('g1', 'figure'),
[Input('g1', 'selectedData'),
Input('g2', 'selectedData'),
Input('g3', 'selectedData')]
)(high_light('Column 0', 'Column 1'))
app.callback(
Output('g2', 'figure'),
[Input('g2', 'selectedData'),
Input('g1', 'selectedData'),
Input('g3', 'selectedData')]
)(high_light('Column 2', 'Column 3'))
app.callback(
Output('g3', 'figure'),
[Input('g3', 'selectedData'),
Input('g1', 'selectedData'),
Input('g2', 'selectedData')]
)(high_light('Column 4', 'Column 5'))
app
(二) 示意图
- 在任一子图表筛选或单击数据,其它子图表会对应更新
-
效果
(三) 说明
- 示例是对由6列数据集,两两组成的3个散点图表,进行交叉筛选的通用示例。对单个散点图进行选择数据区域(或单击),则另外2个散点图会仅显示筛选的数据;
- 对多维数据集进行筛选和可视化,最好选用平行坐标图的方式。
五、Dash交互局限性
- 点击图上的点不能累加:不能累加已经点击的图点数量,也不支持对某个图点进行反选;
- 目前还不能自定义悬停交互及选择框的样式。