Interactive sales dashboard tables
The overarching sales dashboard has come such a long way since you began.
The dashboard has been sent far and wide in the company with some very positive reviews. However, there have been some requests to make the tabular component more interactive. There has been a request to add the ability to select a certain column and have this change the axis on the scatter plot below.
Cet exercice fait partie du cours
Building Dashboards with Dash and Plotly
Instructions
- Enable single row selection in the
grid
by setting therowSelection
parameter below line25
. - Create a callback below line
63
that will use theselectedRows
frommy_grid
to update thescatter_compare
figure with the selected country highlighted as a large red dot.
Exercice interactif pratique
Essayez cet exercice en complétant cet exemple de code.
from dash import Dash, dcc, html, Input, Output, callback
import plotly.express as px
import pandas as pd
from dash_ag_grid import AgGrid
ecom_sales = pd.read_csv('/usr/local/share/datasets/ecom_sales.csv')
logo_link = 'https://assets.datacamp.com/production/repositories/5893/datasets/fdbe0accd2581a0c505dab4b29ebb66cf72a1803/e-comlogo.png'
major_categories = list(ecom_sales['Major Category'].unique())
large_tb = ecom_sales.groupby(['Country'])['OrderValue'].agg(['sum', 'count', 'mean', 'median']).reset_index().rename(columns={'count':'Sales Volume', 'sum':'Total Sales ($)', 'mean':'Average Order Value ($)', 'median':'Median Order Value ($)'})
ecom_country = ecom_sales.groupby('Country')['OrderValue'].agg('sum').reset_index(name='Total Sales ($)')
bar_fig_country = px.bar(ecom_country, x='Total Sales ($)', y='Country', width=500, height=450, title='Total Sales by Country (Hover to filter the Minor Category bar chart!)', custom_data=['Country'], color='Country', color_discrete_map={'United Kingdom':'lightblue', 'Germany':'orange', 'France':'darkblue', 'Australia':'green', 'Hong Kong':'pink'})
money_fmt = {"function": ("params.value.toLocaleString('en-US', {style: 'currency', currency: 'USD'})")}
column_defs = [
{"field": "Country"},
{"field": "Total Sales ($)", "valueFormatter": money_fmt},
{"field": "Average Order Value ($)", "valueFormatter": money_fmt},
{"field": "Median Order Value ($)", "valueFormatter": money_fmt},
{"field": "Sales Volume"},
]
grid = AgGrid(
id="my_grid",
columnDefs=column_defs,
rowData=large_tb.to_dict("records"),
# Enable single-row selection
dashGridOptions={"____": "____"},
defaultColDef={
"cellStyle": {
"textAlign": "left",
"backgroundColor": "black",
"color": "white"}}
)
app = Dash()
app.layout = [
html.Img(src=logo_link,
style={'margin':'30px 0px 0px 0px' }),
html.H1('Sales breakdowns'),
html.Div([
html.H2('Controls'),
html.Br(),
html.H3('Major Category Select'),
dcc.Dropdown(id='major_cat_dd',
options=[{'label':category, 'value':category} for category in major_categories],
style={'width':'200px', 'margin':'0 auto'}),
html.Br(),
html.H3('Minor Category Select'),
dcc.Dropdown(id='minor_cat_dd', style={'width':'200px', 'margin':'0 auto'})],
style={'width':'350px', 'height':'360px', 'display':'inline-block', 'vertical-align':'top', 'border':'1px solid black', 'padding':'20px'}),
html.Div([
html.H3(id='chosen_major_cat_title'),
dcc.Graph(id='sales_line')],
style={'width':'700px', 'height':'380px','display':'inline-block', 'margin-bottom':'5px'}),
grid,
html.Br(),
html.Div([
dcc.Graph(id='scatter_compare'),
dcc.Graph(id='major_cat', figure=bar_fig_country, style={'display':'inline-block'}),
dcc.Graph(id='minor_cat', style={'display':'inline-block'})],
style={'width':'1000px', 'height':'650px','display':'inline-block'})
]
# Create a callback with the right input
@callback(
Output('scatter_compare', 'figure'),
Input('my_grid', '____')
)
def update_scatter(selected_rows):
fig = px.scatter(
large_tb,
x='Sales Volume',
y='Total Sales ($)',
color='Country',
title='Sales Volume vs Total Sales ($) — click a row to highlight')
if selected_rows:
sel_df = pd.DataFrame(selected_rows)
highlight = px.scatter(
sel_df,
x='Sales Volume',
y='Total Sales ($)',
color_discrete_sequence=['#ff0000'],
size=[15] * len(sel_df))
fig.add_trace(highlight.data[0])
return fig
@callback(
Output('minor_cat_dd', 'options'),
Output('chosen_major_cat_title', 'children'),
Input('major_cat_dd', 'value'))
def update_dd(major_cat_dd):
major_minor = ecom_sales[['Major Category', 'Minor Category']].drop_duplicates()
relevant_minor = major_minor[major_minor['Major Category'] == major_cat_dd]['Minor Category'].values.tolist()
minor_options = [dict(label=x, value=x) for x in relevant_minor]
if not major_cat_dd:
major_cat_dd = 'ALL'
major_cat_title = f'This is in the Major Category of : {major_cat_dd}'
return minor_options, major_cat_title
@callback(
Output('sales_line', 'figure'),
Input('minor_cat_dd', 'value'))
def update_line(minor_cat):
minor_cat_title = 'All'
ecom_line = ecom_sales.copy()
if minor_cat:
minor_cat_title = minor_cat
ecom_line = ecom_line[ecom_line['Minor Category'] == minor_cat]
ecom_line = ecom_line.groupby('Year-Month')['OrderValue'].agg('sum').reset_index(name='Total Sales ($)')
line_graph = px.line(ecom_line, x='Year-Month', y='Total Sales ($)', title=f'Total Sales by Month for Minor Category: {minor_cat_title}', height=350)
return line_graph
@callback(
Output('minor_cat', 'figure'),
Input('major_cat', 'hoverData'))
def update_min_cat_hover(hoverData):
hover_country = 'Australia'
if hoverData:
hover_country = hoverData['points'][0]['customdata'][0]
minor_cat_df = ecom_sales[ecom_sales['Country'] == hover_country]
minor_cat_agg = minor_cat_df.groupby('Minor Category')['OrderValue'].agg('sum').reset_index(name='Total Sales ($)')
ecom_bar_minor_cat = px.bar(minor_cat_agg, x='Total Sales ($)', y='Minor Category', orientation='h', height=450, width=480,title=f'Sales by Minor Category for: {hover_country}')
ecom_bar_minor_cat.update_layout({'yaxis':{'dtick':1, 'categoryorder':'total ascending'}, 'title':{'x':0.5}})
return ecom_bar_minor_cat
if __name__ == '__main__':
app.run(debug=True)