Real-time plotly-dash graph from python script

Question

I am trying to create a real-time dash graph using plotly. I am following this tutorial link with the difference that I would like to update the data from another part of the script. In other words, I would like to have my script running and collecting data and then in real time it should be shown on the graph.

This is what I have so far

import dash
from dash.dependencies import Output, Input
import dash_core_components as dcc
import dash_html_components as html
import plotly
import random
import plotly.graph_objs as go
from collections import deque
import pandas as pd


class Plot:

    def __init__(self):
        self.points=[]
        self.app = dash.Dash(__name__)
        self.app.layout = html.Div(
            [
                dcc.Graph(id='live-graph', animate=True),
                dcc.Interval(
                    id='graph-update',
                    interval=500
                ),
            ]
        )
        self. app.callback(Output('live-graph', 'figure'),
        [Input('graph-update', 'n_intervals')])(self.update_graph_scatter)
        
    def start_server(self):
        self.app.run_server(debug=True,host='127.0.0.1', port=8011)

    def update_graph_scatter(self):
        
        pos = pd.DataFrame(data=self.points, columns=['x', 'y'])

        data = plotly.graph_objs.Scatter(
                x=pos['x'],
                y=pos['y'],
                name='Scatter',
                mode= 'lines+markers'
                )

        return {'data': [data],'layout' : go.Layout(xaxis=dict(range=[min(pos['x']),max(pos['x'])]),
                                                    yaxis=dict(range=[min(pos['y']),max(pos['y'])]),)}

    def add_point(self,point):
        self.points.append(point)

What I want is doing something like :

plot=Plot()
plot.start_server()

for i in data_from_my_sensor_callback:
    plot.add_point(i)

But after start_server() the script is not executed. I also tried to run it in a separate thread but the result is the same.

emher · Accepted Answer · 2021-01-05 13:36:34Z

2

To my knowledge, there are presently no official Dash components that support real-time updates initiated from the server, which seems to be what you are after (the Interval component initiates the update from the client). One possible solution would be to use a websocket, which is the approach used by the WebSocket component from dash-extensions. Here is a small example,

import json
import dash_html_components as html
import random
from gevent import sleep
from dash import Dash
from dash.dependencies import Input, Output
from dash_extensions import WebSocket
from dash_extensions.websockets import SocketPool, run_server


# Generator to simulate continuous readings from sensor.
def query_sensor():
    while True:
        sleep(random.uniform(0, 20))  # delay between sensor readings
        yield random.uniform(0, 1)  # the sensor value


# This block runs asynchronously.
def ws_handler(ws):
    for i in query_sensor():
        ws.send(json.dumps(i))  # send data


# Create example app.
app = Dash(prevent_initial_callbacks=True)
socket_pool = SocketPool(app, handler=ws_handler)
app.layout = html.Div([html.Div(id="data"), WebSocket(id="ws")])


@app.callback(Output("data", "children"), [Input("ws", "message")])
def update_data(msg):
    return json.dumps(msg["data"])


if __name__ == '__main__':
    run_server(app, port=5000)  # 5000 if the default port

tested with dash-extensions==0.0.41. Disclaimer: I am the author of dash-extensions.

edited Jan 5, 2021 at 13:36

answered Jan 4, 2021 at 17:54

emher

6,0841 gold badge24 silver badges33 bronze badges

Sign up to request clarification or add additional context in comments.

3 Comments

Polivicio Over a year ago

the example you wrote I think does almost the same thing of the example I am following (see the link in the question). I don't know why but when I run the script nothing happen. Anyway what I would like to do is execute something after the run_server. I also tried to create a separate thread but the server block the execution of the script.

emher Over a year ago

If you go to localhost:5000 (the default port), you should see a number (representing the data) updating. As you note, you can’t just put code after run_server. However, the ws_handler is executed asynchronously, and should thus yield the desired behavior if you put the data collection code there.

emher Over a year ago

Great! I have updated the code to better fit your use case (according to the question update).

Collectives™ on Stack Overflow

Real-time plotly-dash graph from python script

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related