使用 Matplotlib 中的数字的动态子图

Question

I want to create a subplot using matplotlib with pysimplegui where each time i select some signals-checkbox(list of values) i should get a plot corresponding to it which will dynamically increase or decrease in size based on my selection so when i select a checkbox the respective plot will be plotted and eqaully spaced and when i deselect a plot it should be gone and other plots should automatically occupy the space Programming Language - Python3 Matplotlib - 3.6.2

My Requirement:

1. There will always be maximum 10 rows and 1 column in my subplot (matplotlib)
2. So lets say iam making a call to "plt.show()" after making 1 plot my result should have only one plot at that moment occupying the whole figure
3. Now iam adding one more plot and then calling "plt.show()", now it should have 2 plots occupying the fig equally this would be the same for 3,4,5... 10 plots but all these plots should have a single shared axis whichever is at the bottom
4. And also a way to delete a plot at any instance of time so for instance i make a plot say axis1 and then I make a plot axis2 now I want axis1 to be deleted after which only my axis2 plot will be there occupying the whole figure

The below code satisfies most of my requirement but its just that the axis are not shared with each other and also the sizing of each plots are different

reff1 : Changing matplotlib subplot size/position after axes creation reff2 : Dynamically add/create subplots in matplotlib

import matplotlib.pyplot as plt
import matplotlib.gridspec as gridspec
 
number_of_plots = 2
fig = plt.figure()
 
ax = fig.add_subplot(1, 1, 1)
 
gs = gridspec.GridSpec(number_of_plots + 1, 1)
plot_space = gs[0:number_of_plots].get_position(fig)
print(plot_space,
      [plot_space.x0, plot_space.y0, plot_space.width, plot_space.height])
ax.set_position(plot_space)
ax.set_subplotspec(gs[0:number_of_plots])              # only necessary if using tight_layout()
fig.tight_layout()                # not strictly part of the question
ax = fig.add_subplot(gs[2], sharex=ax)
 
plt.show()

Answer 1

Here is a basis for the interactive part.

I remove all axes from the figure at each new selection and add the newly selected ones on a relevant GridSpec .

Still axes are not redrawn every time thanks to @lru_cache .

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.gridspec import GridSpec
from functools import lru_cache

plt.ion()


def main():

    # Data that will be selected
    datas = {
        "data1": np.random.randn(100),
        "data2": np.random.randn(100),
        "data3": np.random.randn(100),
        "data4": np.random.randn(100),
    }

    # Emulate data selection over time
    states = (
        ["data1"],
        ["data1", "data3"],
        ["data2", "data3"],
        ["data1", "data2", "data3"],
        ["data2"],
    )

    fig = plt.figure(figsize=(6, 8))

    # Create a new ax on fig with corresponding data plotted
    @lru_cache
    def get_data_ax(data_id):
        ax = fig.add_subplot()
        ax.plot(datas[data_id])
        ax.set_title(data_id)
        return ax

    # Emulate data selection over time
    for data_state in states:

        # Remove all axes, add relevant ones later
        for ax in fig.axes:
            fig.delaxes(ax)

        # Create a GridSpec for the axes to be plot
        num_plots = len(data_state)
        # but give it at least 2 rows so a single ax will not be stretched
        num_rows = max(2, num_plots)
        gs = GridSpec(num_rows, 1)

        # Get all axes to be plotted
        # (thanks to @lru_cache, axes are computed only once)
        axes = [get_data_ax(data_id) for data_id in data_state]

        # Add each ax one by one
        for ax_idx, ax in enumerate(axes):
            # Choose the right location for the given ax
            ax.set_subplotspec(gs[ax_idx])
            fig.add_subplot(ax)

        plt.pause(2)


if __name__ == "__main__":
    main()