ui.py

import dash
from dash.dependencies import Input, Output, State
import dash_core_components as dcc
import dash_html_components as html
import dash_bootstrap_components as dbc
import tabs as tb
import datetime 
# imports backend
import server as sv

# Uses bootstrap stylesheet
external_stylesheets = [dbc.themes.BOOTSTRAP]

app = dash.Dash(__name__, external_stylesheets=external_stylesheets)

##Creates layout

app.layout = html.Div([
    html.Div([html.Img(src='/static/images/databio_logo.svg',
        style = {'display': 'inline', 'width': '50%', 'max-width': '8rem'}),
        html.H1("Covid19 Analysis 🇨🇱 V0.1", style={'text-align': 'center', 'color':'white'}), 
        html.P(['made with ❤ by ', html.A('millacurafa', href='https://github.com/millacurafa', style={'color':'white'})],
        style={'text-align': 'center', 'color':'white'}),
        ], className = 'navbar navbar-primary bg-dark'),
    dcc.Tabs(id='tabs_chosen', value='tab-1', children=[
        dcc.Tab(label='National', value='tab-1'),
        dcc.Tab(label='Regional', value='tab-2'),
       # dcc.Tab(label='SEIRD real data', value='tab-3'),
        dcc.Tab(label='SEIRD model', value='tab-4'),
        dcc.Tab(label='Docs', value='tab-5'),
    ]),
    html.Div(id='tabs_content')
])

##Generates callbacks
@app.callback(        
    Output('tabs_content', 'children'),
    Input('tabs_chosen', 'value')
)
def render_content(tab):
    if tab == 'tab-1':
        return html.Div(
            tb.tab_1
        )
    elif tab == 'tab-2':
        return html.Div(
            tb.tab_2
        )
  #  elif tab == 'tab-3':
   #     return html.Div(    
    #        tb.tab_3
     #   )
    elif tab == 'tab-4':
        return html.Div(
            tb.tab_4
        )
    elif tab == 'tab-5':
        return html.Div(
            tb.tab_5
        )

@app.callback(
 Output('time_series_one', 'figure'),
 Input('submit_button_state_one', 'n_clicks'),
[
    ##For tab_1
    State('national_dropdown', 'value'),
    State('national_switches_input', 'value'),
    State('national_datepicker', 'start_date'),
    State('national_datepicker', 'end_date')
           
])
def update_figure(_, national_dropdown,national_switches_input, start_date, end_date):
    dff = sv.df ##Creates a copy of the dataframe
    [national_dropdown if national_dropdown != None else 'Casos totales']
    dff = dff.loc[(dff.index >= start_date) & (dff.index <= end_date), national_dropdown]
    n_switches = sv.np.sum(national_switches_input)
    fig = sv.px.line(dff).update_layout(title= "National cases",
                      yaxis_title='Number of National cases',
                      xaxis_title='Date')
    if n_switches != 0:
        if n_switches==1:
            dff = dff/(sv.N/1000)
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases per 1000 inhabitants',
                      xaxis_title='Date')
            return fig
        elif n_switches==2:
            dff = sv.np.log10(dff)
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases (in log10 scale)',
                      xaxis_title='Date')
            return fig
        elif n_switches==3:
            dff = sv.np.log10(dff/(sv.N/1000))
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases per 1000 inhabitants (in log10 scale)',
                      xaxis_title='Date')
            return fig
        elif n_switches==4:
            dff = dff.rolling(14).mean()
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases (MA14)',
                      xaxis_title='Date')
            return fig
        elif n_switches==5:
            dff = (dff/(sv.N/1000)).rolling(14).mean()
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases per 1000 inhabitants (MA14)',
                      xaxis_title='Date')
            return fig
        elif n_switches==6:
            dff = sv.np.log10(dff.rolling(14).mean())
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases (log10(MA14))',
                      xaxis_title='Date')
            return fig
        elif n_switches==7:
            dff = sv.np.log10((dff/(sv.N/1000)).rolling(14).mean())
            fig = sv.px.line(dff)
            fig.update_layout(title= "National cases",
                      yaxis_title='Number of National cases per 1000 inhabitants (log10(MA14))',
                      xaxis_title='Date')
            return fig
    else: return fig 


@app.callback(

        Output('time_series_two', 'figure'),
    
        Input('submit_button_state_two', 'n_clicks'),
   
    [# For tab_2
            State('regional_dropdown', 'value'),
            State('regional_cases', 'value'),
            State('regional_switches_input', 'value'),
            State('regional_datepicker', 'start_date'),
            State('regional_datepicker', 'end_date')
    ])

def regional(_,regional_dropdown,regional_cases,regional_switches_input,start_date,end_date):
    n_switches = sv.np.sum(regional_switches_input)
    if (regional_cases == 'active'):
        if (n_switches==1 or n_switches==3):
            dff = sv.df_region_current_bypop['bypop']
        else:
            dff = sv.df_region_current
    elif regional_cases == 'total':
        if (n_switches==1 or n_switches==3):
            dff = sv.df_region_current_bypop['bypop'].cumsum()
        else:
            dff = sv.df_region_current.cumsum()
    elif regional_cases == 'deaths':
        if (n_switches==1 or n_switches==3):
            dff = sv.df_deaths_current_bypop['bypop']
        else:
            dff = sv.df_deaths_current
    elif regional_cases == 'uci':
        if (n_switches==1 or n_switches==3):
            dff = sv.df_region_current_bypop['bypop']
        else:
            dff = sv.df_uci_current   
    elif regional_cases == 'pcr':
        if (n_switches==1 or n_switches==3):
            dff = sv.df_region_current_bypop['bypop']
        else:
            dff = sv.df_pcr_current

    dff = dff.loc[(dff.index >= start_date) & (dff.index <= end_date),]
    dff = dff.filter(regional_dropdown, axis=1)
    fig = sv.px.line(dff
            ).update_layout(title= "Regional cases",
                      yaxis_title='Number of cases by region',
                      xaxis_title='Date')

    if n_switches != 0:
        if (n_switches==2 or n_switches==3):
            dff = sv.np.log10(dff)
            fig = sv.px.line(dff).update_layout(title= "Regional cases",
                      yaxis_title='Number of cases by region (in log10 scale)',
                      xaxis_title='Date')
            return fig
        else:
            return fig
    else: return fig



@app.callback(
    Output('time_series_three', 'figure'),
    Input('submit_button_state_three', 'n_clicks'),
    [
            ##For tab_3
            State('seird_dropdown', 'value'),
            State('seird_datepicker', 'start_date'),
            State('seird_datepicker', 'end_date')
    ])
    
def plotrealseird(_, seird_dropdown,start_date,end_date):
    dff = sv.df_seird
    [seird_dropdown if seird_dropdown != None else ['Susceptible','Exposed','Infectious','Recovered','Deaths']]
    dff = dff.loc[(dff.index >= start_date) & (dff.index <= end_date),seird_dropdown]
    fig = sv.px.line(dff
            ).update_layout(title='SEIRD model real data',
                      yaxis_title='SEIRD cases',
                      xaxis_title='Date')
    return fig

# @app.callback(
#     [
     
#         Output('seird_city_dropdown', 'options'),
#         #Output('seirdmo_population', 'value')
#     ],
#     [
     
#         Input('seird_regional_dropdown','value'),
#         #Input('seird_city_dropdown','value')
    
#     ])

# def updatedropdown(seird_regional_dropdown):
#     cities = sv.df_city_current[sv.df_city_current['Region']== seird_regional_dropdown].groupby('Comuna')[['Poblacion']].mean()
#     OptionList = [{'label': city, 'value': city} for city in cities.index]
#     OptionList.insert(0,{'label': 'Total', 'value': 'Total'})
#     return OptionList


@app.callback(
    
            Output('time_series_four', 'figure'),
            #Output('seirdmo_days_today', 'children')     

            Input('submit_button_state_four', 'n_clicks'),
    [
            ##For tab_4
            State('seirdmo_daypicker', 'date'),
            State('seirdmo_days_today', 'value'),
            State('seirdmo_initial_cases', 'value'),
            State('seirdmo_initial_deaths', 'value'),
            State('seirdmo_initial_exposed', 'value'),
            State('seirdmo_initial_recovered', 'value'),            
            State('seirdmo_population', 'value'),
            State('seirdmo_icu_beds', 'value'),
            State('seirdmo_p_I_to_C', 'value'),
            State('seirdmo_p_C_to_D', 'value'),
            State('seirdmo_r0_slider', 'value'),

    ])

def plotseirdgo(_, 
                date, 
                seirdmo_days_today,
                seirdmo_initial_cases,
                seirdmo_initial_deaths,
                seirdmo_initial_exposed,
                seirdmo_initial_recovered,
                seirdmo_population,
                seirdmo_icu_beds,
                seirdmo_p_I_to_C,
                seirdmo_p_C_to_D,
                seirdmo_r0_slider):
    
    # if date != '2020-01-01':
    #     delta = (datetime.datetime.today() - datetime.datetime.strptime(date,'%Y-%m-%d')).days
    # else:
    #     delta = (datetime.datetime.today() - datetime.datetime(2020,1,1)).days
        
    N = seirdmo_population #Chilean population; Source: World Bank
    deaths_model = seirdmo_initial_deaths #deaths
    recovered_model = seirdmo_initial_recovered #recovered
    infectious_model = seirdmo_initial_cases #infectious
    susceptible_model = N -1 #susceptible
    exposed_model = seirdmo_initial_exposed # contracted the disease but are not yet infectious 
    D = 10 # Infectious lasts
    gamma = 1/D
    R0 = seirdmo_r0_slider # the total number of people an infected person infects
    beta = R0*gamma # infected person infects beta people per day
    alpha = seirdmo_p_C_to_D/100 # five percent death rate
    rho = 1/14 # fourteen days from infection until death
    delta = 1/7  # incubation period of seven days
    S0, E0, I0, R0, D0 = susceptible_model, exposed_model, infectious_model, R0, deaths_model  #Initial conditions

    ##Creates time
    t = sv.np.linspace(0, seirdmo_days_today) # Grid of time points (in days)
    y0 = S0, E0, I0, R0, D0 # Initial conditions vector

    # Integrate the SIR equations over the time grid, t.
    ret = sv.odeint(sv.derivate, y0, t, args=(N, beta, gamma, delta, alpha, rho))
    S, E, I, R, D = ret.T
    return sv.plotlyseirdgo(t, S, E, I, R, D)
      

server = app.server
app.config.suppress_callback_exceptions = True



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