JupyterDash in Colab - very slow

Dash Python
1

I’ve been playing with JuypterDash in Colab. It’s a fantastic solution, but… it seems so slow!
My layout is a super-simple form with one dcc.Input field and 1 button, used to collect usernames in a teaching environment.

I run it in ‘external’ mode to allow trainees to open the page from their computer

Just to open the form takes upward of 30 seconds (during which “Loading” is showing). I don’t quite understand what makes it so slow…

Anyone with a recommendation?

from jupyter_dash import JupyterDash
import dash_core_components as dcc
import dash_html_components as html
from dash.dependencies import Input, Output, State
import pandas as pd

app = JupyterDash(__name__)
db = []

# Define layout
app.layout = html.Div([
    html.H1("Registration"),
    html.Label([
        "Your email ",
        dcc.Input(
            id='account-email'
        )
    ]),
    html.Button('Submit', id='submit-val', n_clicks=0),
    html.P(
        id='feedback-output'
    )
])

# Define callback to update graph
@app.callback(
    Output('feedback-output', 'children'),
    Input("submit-val", "n_clicks"),
    State("account-email", "value")
)
def update_output(submit, email):
    if email:
        db.append(dict(email=email))
    return email


# Run app and display result inline in the notebook
app.run_server(mode='external')