Py.Cafe

import dash
from dash import dcc, html, Input, Output
import dash_bootstrap_components as dbc
import pandas as pd
import numpy as np
import plotly.figure_factory as ff
import plotly.colors as pc
import plotly.graph_objects as go


def genSankey(df, cat_cols=[], value_cols='', c_scale='', region ='', year='', shipmode=''):
    labelList = []
    colorNumList = []
    for catCol in cat_cols:
        labelListTemp = list(set(df[catCol].values))
        colorNumList.append(len(labelListTemp))
        labelList = labelList + labelListTemp

    # Remove duplicates from labelList
    labelList = list(dict.fromkeys(labelList))

    # Define base colors using Viridis color scale
    num_labels = len(labelList)
    color_scale = getattr(pc.sequential, c_scale)
    base_colors = pc.sample_colorscale(color_scale, [i / (num_labels - 1) for i in range(num_labels)])
    
    # Assign colors directly
    color_dict = {label: color for label, color in zip(labelList, base_colors)}

    # Transform df into a source-target pair
    for i in range(len(cat_cols) - 1):
        if i == 0:
            sourceTargetDf = df[[cat_cols[i], cat_cols[i + 1], value_cols]]
            sourceTargetDf.columns = ['source', 'target', 'count']
        else:
            tempDf = df[[cat_cols[i], cat_cols[i + 1], value_cols]]
            tempDf.columns = ['source', 'target', 'count']
            sourceTargetDf = pd.concat([sourceTargetDf, tempDf])
        
        sourceTargetDf = sourceTargetDf.groupby(['source', 'target']).agg({'count': lambda x: round(x.sum(), 2)}).reset_index()

    # Add index for source-target pair
    sourceTargetDf['sourceID'] = sourceTargetDf['source'].apply(lambda x: labelList.index(x))
    sourceTargetDf['targetID'] = sourceTargetDf['target'].apply(lambda x: labelList.index(x))

    # Create a list of colors for the links with alpha 0.5
    link_colors = [color_dict[src].replace('rgb(', 'rgba(').replace(')', ', 0.5)') for src in sourceTargetDf['source']]

    # Creating the sankey diagram
    data = dict(
        type='sankey',
        node=dict(
            pad=15,
            thickness=20,
            line=dict(
                color="black",
                width=0.5
            ),
            label=labelList,
            color=[color_dict[label] for label in labelList]
        ),
        link=dict(
            source=sourceTargetDf['sourceID'],
            target=sourceTargetDf['targetID'],
            value=sourceTargetDf['count'].apply(lambda x: float("{:.2f}".format(x))),
            color=link_colors
        )
    )

    layout = dict(template='seaborn',
                  title=dict(
                  text=f'Sankey of {"all years" if year == "All" else year} sales data, '
                        f'for {"all regions" if region == "All" else "the " + region.lower() + " region"}, '
                        f'and {"all shipping modes" if shipmode == "All" else shipmode.lower() + " shipping mode"}',
                    ),
                    width=900,
                    height=500,
                    font=dict(color='#FFFFFF'),
                    margin=dict(b=35, t=35, l=0, r=0),
                    paper_bgcolor='rgba(0,0,0,0)',
                    plot_bgcolor='rgba(0,0,0,0)'
                  
    )

    fig = dict(data=[data], layout=layout)

    return fig


def sales_map(df,selected_region, selected_year, selected_shipmode):
    fig = ff.create_hexbin_mapbox(
        data_frame=df, lat="Lat", lon="Long",
        nx_hexagon=20,
        opacity=0.6,
        color="Sales", 
        agg_func=np.sum,
        labels={"color": "Sales"}, 
        mapbox_style='open-street-map', 
        min_count=1, 
        color_continuous_scale='plasma',
    )

    fig.update_layout(title=dict(
                    text=f'Map of {"all years" if selected_year == "All" else selected_year} sales data, '
                        f'for {"all regions" if selected_region == "All" else "the " + selected_region.lower() + " region"}, '
                        f'and {"all shipping modes" if selected_shipmode == "All" else selected_shipmode.lower() + " shipping mode"}',
                    ),
                    width=900,
                    height=500,
                    font=dict(color='#FFFFFF'),
                    margin=dict(b=35, t=35, l=0, r=0),
                    paper_bgcolor='rgba(0,0,0,0)',
                    plot_bgcolor='rgba(0,0,0,0)'
                    )

    return fig


def sankey_figure(df, selected_region, selected_year, selected_shipmode):
    grouped_df = df.groupby(['Segment','State/Province','Category']).agg({
        'Sales': 'sum',
    #     'Profit': 'sum'
    }).reset_index()

    cat_columns = grouped_df.columns.tolist()
    x = len(cat_columns)-1
    cat_columns.pop(x)
    cat_columns

    fig = genSankey(df, cat_cols=cat_columns,value_cols='Sales',c_scale='Darkmint', region=selected_region, year=selected_year, shipmode=selected_shipmode)  

    return fig


# Load data
data_df = pd.read_csv('Superstore.csv')
data_df['Order Date'] = pd.to_datetime(data_df['Order Date'], dayfirst=True)
data_df['order_year'] = data_df['Order Date'].dt.year

# Extract unique values for filters
regions = sorted(data_df['Region'].unique())
regions.insert(0, 'All')
orderyears = sorted(data_df['order_year'].unique(), reverse=True)
orderyear_options = [{'label': str(orderyear), 'value': orderyear} for orderyear in orderyears]
orderyear_options.insert(0, {'label': 'All', 'value': 'All'})

shipmodes = sorted(data_df['Ship Mode'].unique())
shipmodes.insert(0, 'All')

# Initialize the Dash app
app = dash.Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])

# Define custom theme
custom_theme = {
    "primary": "#75C9BE",
    "secondary": "#d6e8e7",
    "info": "#3B9ECB",
    "gray": "#adb5bd",
    "success": "#8BE3AA",
    "warning": "#F9F871",
    "danger": "#c0003e",
    "body_bg": "#1F5869",
    "content_bg": "#153F4C",
    "text_color": "#CBE2E2"
}

# Define the layout
app.layout = dbc.Container(
    [
        dbc.Row(
            dbc.Col(
                html.H1(
                    'Figure Friday - Week 28 - Sample Superstore Data',
                    style={'textAlign': 'center', 'color': custom_theme["text_color"]}
                ),
                width=12
            ),
            justify='center'
        ),
        dbc.Row(
            [
                dbc.Col(
                    [
                        html.Label('Select a Region', style={'color': custom_theme["text_color"]}),
                        dcc.Dropdown(id='region-filter', options=[{'label': region, 'value': region} for region in regions], value='All')
                    ],
                    width=4,
                    className="my-4",
                    style={'padding': '0 5px'}
                ),
                dbc.Col(
                    [
                        html.Label('Select an Order Year', style={'color': custom_theme["text_color"]}),
                        dcc.Dropdown(id='sales_year-filter', options=orderyear_options, value='All')
                    ],
                    width=4,
                    className="my-4",
                    style={'padding': '0 5px'}
                ),
                dbc.Col(
                    [
                        html.Label('Select a Shipping Mode', style={'color': custom_theme["text_color"]}),
                        dcc.Dropdown(id='ship_mode-filter', options=[{'label': shipmode, 'value': shipmode} for shipmode in shipmodes], value='All')
                    ],
                    width=4,
                    className="my-4",
                    style={'padding': '0 5px'}
                ),
            ],
            justify='center',
            style={'margin-bottom': '20px'}
        ),
        dbc.Row(
            [
                dbc.Col(
                    dcc.Graph(id='sales_graph1'),
                    width=6,
                    className="my-4",
                    style={'padding': '0 5px'}
                ),
                dbc.Col(
                    dcc.Graph(id='sankey'),
                    width=6,
                    className="my-4",
                    style={'padding': '0 5px'}
                ),
            ]
        )
    ],
    fluid=True,
    style={'backgroundColor': custom_theme["body_bg"]}
)



# Define callbacks to update figures based on selected filters
@app.callback(
    [Output('sales_graph1', 'figure'),
     Output('sankey', 'figure')],
    [Input('region-filter', 'value'),
     Input('sales_year-filter', 'value'),
     Input('ship_mode-filter', 'value')]
)
def update_figures(selected_region, selected_year, selected_shipmode):
    filtered_df = data_df
    
    if selected_region != 'All':
        filtered_df = filtered_df[filtered_df['Region'] == selected_region]
    
    if selected_year != 'All':
        filtered_df = filtered_df[filtered_df['order_year'] == selected_year]
    
    if selected_shipmode != 'All':
        filtered_df = filtered_df[filtered_df['Ship Mode'] == selected_shipmode]
    
    sales_map_fig = sales_map(filtered_df, selected_region, selected_year, selected_shipmode)
    sankey_fig = sankey_figure(filtered_df, selected_region, selected_year, selected_shipmode)
    
    return sales_map_fig, sankey_fig


if __name__ == '__main__':
    app.run_server(debug=True)