Live Plot in Python GUI

Question

I am trying to write a Python GUI and I need to do a live plot. I currently have a program that receives data from a machine I am using and I want to be able to plot the val

Answer 1

It looks like you get the data by polling, which means you don't need threads or multiple processes. Simply poll the device at your preferred interface and plot a single point.

Here's an example with some simulated data to illustrate the general idea. It updates the screen every 100ms.

import Tkinter as tk
import random

class ServoDrive(object):
    # simulate values
    def getVelocity(self): return random.randint(0,50)
    def getTorque(self): return random.randint(50,100)

class Example(tk.Frame):
    def __init__(self, *args, **kwargs):
        tk.Frame.__init__(self, *args, **kwargs)
        self.servo = ServoDrive()
        self.canvas = tk.Canvas(self, background="black")
        self.canvas.pack(side="top", fill="both", expand=True)

        # create lines for velocity and torque
        self.velocity_line = self.canvas.create_line(0,0,0,0, fill="red")
        self.torque_line = self.canvas.create_line(0,0,0,0, fill="blue")

        # start the update process
        self.update_plot()

    def update_plot(self):
        v = self.servo.getVelocity()
        t = self.servo.getTorque()
        self.add_point(self.velocity_line, v)
        self.add_point(self.torque_line, t)
        self.canvas.xview_moveto(1.0)
        self.after(100, self.update_plot)

    def add_point(self, line, y):
        coords = self.canvas.coords(line)
        x = coords[-2] + 1
        coords.append(x)
        coords.append(y)
        coords = coords[-200:] # keep # of points to a manageable size
        self.canvas.coords(line, *coords)
        self.canvas.configure(scrollregion=self.canvas.bbox("all"))

if __name__ == "__main__":
    root = tk.Tk()
    Example(root).pack(side="top", fill="both", expand=True)
    root.mainloop()

Answer 2

As far as I understand it, it can be done with tkinter/Qt/wxpython etc. You just have to make use of mutlithreading and multiprocessing. There may be a simpler way to do it with another module, but I am unaware of it.

I have been looking into something similar to this problem for a long time as well, and it appears that it is a constant issue among this community.

Here are some threads which talk about the issue:

How do I refresh a matplotlib plot in a Tkinter window?

How do I update a matplotlib figure while fitting a function?