PySimpleGUI - Matplotlib Integration

When Matplotlib is used from Python shell, the plots are displayed in a default window. The backend_tkagg module is useful for embedding plots in Tkinter.

The Canvas element in PySimpleGUI has TKCanvas method that returns original TKinter's Canvas object. It is given to the FigureCanvasTkAgg() function in the backend_tkagg module to draw the figure.

First, we need to create the figure object using the Figure() class and a plot to it. We shall draw a simple plot showing sine wave.

fig = matplotlib.figure.Figure(figsize=(5, 4), dpi=100)
t = np.arange(0, 3, .01)
fig.add_subplot(111).plot(t, 2 * np.sin(2 * np.pi * t))

Define a function to draw the matplotlib figure object on the canvas

def draw_figure(canvas, figure):
   figure_canvas_agg = FigureCanvasTkAgg(figure, canvas)
   figure_canvas_agg.draw()
   figure_canvas_agg.get_tk_widget().pack(side='top', fill='both', expand=1)
   return figure_canvas_agg

Obtain the Canvas from PySimpleGUI.Canvas object by calling its TkCanvas property.

layout = [
   [psg.Text('Plot test')],
   [psg.Canvas(key='-CANVAS-')],
   [psg.Button('Ok')]
]

Draw the figure by calling the above function. Pass the Canvas object and fifure object to it.

fig_canvas_agg = draw_figure(window['-CANVAS-'].TKCanvas, fig)

Example: Draw a Sinewave Line graph

The complete code is given below −

import matplotlib.pyplot as plt
import numpy as np
from matplotlib.backends.backend_tkagg import
FigureCanvasTkAgg
import PySimpleGUI as sg
import matplotlib
matplotlib.use('TkAgg')
fig = matplotlib.figure.Figure(figsize=(5, 4), dpi=100)
t = np.arange(0, 3, .01)
fig.add_subplot(111).plot(t, 2 * np.sin(2 * np.pi * t))
def draw_figure(canvas, figure):
   tkcanvas = FigureCanvasTkAgg(figure, canvas)
   tkcanvas.draw()
   tkcanvas.get_tk_widget().pack(side='top', fill='both', expand=1)
   return tkcanvas
layout = [[sg.Text('Plot test')],
   [sg.Canvas(key='-CANVAS-')],
   [sg.Button('Ok')]]
window = sg.Window('Matplotlib In PySimpleGUI', layout, size=(715, 500), finalize=True, element_justification='center', font='Helvetica 18')

# add the plot to the window
tkcanvas = draw_figure(window['-CANVAS-'].TKCanvas, fig)
event, values = window.read()
window.close()

The generated graph is as follows −