make_figures.py

import numpy as np
import pandas as pd

import dash_core_components as dcc
import dash_table
from dash_table.Format import Format
import dash_table.FormatTemplate as FormatTemplate
import plotly.graph_objects as go

# attendance summary
def make_attendance_table(df):
    # check if not enough info
    if df.loc[0, 'attendance_category'] == 'Not enough info':
        sure_bet_count, at_risk_count, not_met_count, on_track_count = [None] * 4
        sure_bet_pct, at_risk_pct, not_met_pct, on_track_pct = [None] * 4
    else:
        # count children in attendance category
        # using shape[0] to count rows and return an int
        sure_bet_count = df[df['attendance_category'] == 'Sure bet'].shape[0]
        at_risk_count = df[df['attendance_category'] == 'At risk'].shape[0]
        not_met_count = df[df['attendance_category'] == 'Not met'].shape[0]
        on_track_count = df[df['attendance_category'] == 'On track'].shape[0]
        total_count = df.loc[df['attendance_category'] != 'Case expired'].shape[0]

        # calculate percentage of children in each category
        sure_bet_pct = sure_bet_count / total_count
        at_risk_pct = at_risk_count / total_count
        not_met_pct = not_met_count / total_count
        on_track_pct = on_track_count / total_count

    # make table
    label_col = [
        'Sure bet',
        'On track',
        'At risk',
        'Not met'
    ]
    pct_col = [
        sure_bet_pct,
        on_track_pct,
        at_risk_pct,
        not_met_pct
    ]
    count_col = [
        sure_bet_count,
        on_track_count,
        at_risk_count,
        not_met_count
    ]
    df_table = pd.DataFrame(
        {
            'attendance_category': label_col,
            'percentage': pct_col,
            'count': count_col
        }
    )

    table = dash_table.DataTable(
        id='summary',
        columns=[
            {
                'id': 'attendance_category',
                'name': 'Attendance risk',
                'type': 'text'
            }, {
                'id': 'percentage',
                'name': 'Percentage',
                'type': 'numeric',
                'format': {'nully':'-%',
                        'prefix': None,
                        'specifier': '.0%'}
            }, {
                'id': 'count',
                'name': 'Count',
                'type': 'numeric',
                'format': Format(
                    precision=0,
                    nully='-'
                )
            }
        ],
        data=df_table.to_dict('records'),
        style_as_list_view=True,
        style_table={
            'overflowX': 'auto',
            'padding':'20px'
        },
        style_data_conditional=[
            {
                'if': {
                    'filter_query': '{attendance_category} = "Sure bet"',
                    'column_id': 'attendance_category'
                },
                'color': '#0f8c48'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "Not met"',
                    'column_id': 'attendance_category'
                },
                'color': '#da5d4f'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "At risk"',
                    'column_id': 'attendance_category'
                },
                'color': '#da5d4f'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "On track"',
                    'column_id': 'attendance_category'
                },
                'color': '#0f8c48'
            }
        ]
    )
    return table

# revenue barchart
def make_revenue_chart(df):
    # sum up revenues
    min_revenue_sum = df['min_revenue'].sum()
    potential_revenue_sum = df['potential_revenue'].sum()
    max_approved_revenue_sum = df['max_revenue'].sum()
    potential_e_learning_revenue_sum = df['e_learning_revenue_potential'].sum()

    min_potential_delta = potential_revenue_sum - min_revenue_sum
    potential_max_delta = max_approved_revenue_sum - potential_revenue_sum

    trace_min = go.Bar(
        name=(
            'Guaranteed revenue <br>' + '$' + str(round(min_revenue_sum))
        ),
        y=['revenue'],
        x=[min_revenue_sum],
        width=[0.3],
        orientation='h',
        marker_color='rgb(0,110,160)'
    )
    trace_potential = go.Bar(
        name=(
            'Potential revenue <br>' + '$' + str(round(potential_revenue_sum))
        ),
        y=['revenue'],
        x=[min_potential_delta],
        width=[0.3],
        orientation='h',
        marker_color='rgb(56,178,234)'
    )
    trace_max = go.Bar(
        name=(
            'Max. approved revenue <br>' + '$' + str(round(max_approved_revenue_sum))
        ),
        y=['revenue'],
        x=[potential_max_delta],
        width=[0.3],
        orientation='h',
        marker_color='rgb(204,239,255)'
    )
    trace_elearning = go.Bar(
        name=(
            'Potential e-learning revenue <br>'
            + '$' + str(round(potential_e_learning_revenue_sum))
        ),
        y=['revenue'],
        x=[potential_e_learning_revenue_sum],
        width=[0.3],
        orientation='h',
        marker_color='rgb(230,230,230)',
    )

    data = [trace_min, trace_potential, trace_max, trace_elearning]
    layout = go.Layout(
        barmode='stack',
        yaxis={
            'visible':False,
            'showticklabels':False,
            'fixedrange':True
        },
        xaxis={
            'showgrid':False,
            'visible':False,
            'showticklabels':False,
            'fixedrange':True,
        },
        plot_bgcolor='rgb(255,255,255)',
        margin={
            'l':0,
            'r':0,
            't':0,
            'b':0,
        },
        legend={
            'orientation':'h',
            'traceorder':'normal'
        },
        hovermode=False,
        font={'size':16},
        height=200,
    )
    fig = go.Figure(data=data, layout=layout)
    return dcc.Graph(
        figure=fig,
        config={'displayModeBar':False},
    )


# child level table
def make_table(df):
    table = dash_table.DataTable(
        id='child_level',
        data=df.to_dict('records'),
        columns=[
            {
                'id': 'name',
                'name': 'Child name',
                'type': 'text'
            }, {
                'id': 'case_number',
                'name': 'Case number',
                'type': 'text'
            }, {
                'id': 'biz_name',
                'name': 'Business',
                'type': 'text'
            }, {
                'id': 'attendance_category',
                'name': 'Attendance risk',
                'type': 'text'
            }, {
                'id': 'attendance_rate',
                'name': 'Attendance rate',
                'type': 'numeric',
                'format': {'nully':'-%',
                        'prefix': None,
                        'specifier': '.0%'}
            }, {
                'id': 'min_revenue',
                'name': 'Guaranteed revenue',
                'type': 'numeric',
                'format': FormatTemplate.money(2)
            }, {
                'id': 'potential_revenue',
                'name': 'Potential revenue',
                'type': 'numeric',
                'format': FormatTemplate.money(2)
            }, {
                'id': 'max_revenue',
                'name': 'Max. approved revenue',
                'type': 'numeric',
                'format': FormatTemplate.money(2)
            }, {
                'id': 'e_learning_revenue_potential',
                'name': 'Potential e-learning revenue',
                'type': 'numeric',
                'format': FormatTemplate.money(2)
            }
        ],
        style_data_conditional=[
            {
                'if': {
                    'filter_query': '{attendance_category} = "Not enough info"',
                    'column_id': 'attendance_category'
                },
                'backgroundColor': 'rgb(248, 248, 248)',
                'color': 'rgb(128,128,128)'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "Sure bet"',
                    'column_id': 'attendance_category'
                },
                'color': '#0f8c48'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "Sure bet"',
                    'column_id': 'attendance_rate'
                },
                'color': '#0f8c48'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "Not met"',
                    'column_id': 'attendance_category'
                },
                'color': '#da5d4f'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "Not met"',
                    'column_id': 'attendance_rate'
                },
                'color': '#da5d4f'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "At risk"',
                    'column_id': 'attendance_category'
                },
                'color': '#da5d4f'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "At risk"',
                    'column_id': 'attendance_rate'
                },
                'color': '#da5d4f'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "On track"',
                    'column_id': 'attendance_category'
                },
                'color': '#0f8c48'
            },
            {
                'if': {
                    'filter_query': '{attendance_category} = "On track"',
                    'column_id': 'attendance_rate'
                },
                'color': '#0f8c48'
            },
        ],
        style_header={
            'whiteSpace': 'normal',
            'height': 'auto',
            'maxWidth': '100px',
        },
        style_table={
            'padding': '20px',
            'overflowX': 'auto',
        },
        sort_action='native',
        sort_mode='single',
    )
    return table

if __name__ == '__main__':
    from data_input import get_dashboard_data
    df_dashboard, latest_date, is_data_insufficient, days_req_for_warnings = get_dashboard_data()

    # fig = make_revenue_chart(df_dashboard)
    # fig.show()

    make_attendance_table(df_dashboard)