PySimpleGUI，在一个图中生成汇总统计和直方图表

Question

我在 Python 中遇到了问题。请看下面的代码。它在控制台中生成汇总统计信息和直方图。但是，我想使用 PySimpleGUI 创建一个 GUI，它将提供一个 GUI，允许用户更改绘图的参数，因此每次都会生成新的直方图和汇总统计信息。我对如何进行此操作一无所知，因为我对 Python 和 Stack Overflow 还是很陌生。我希望在运行代码时出现一个用户界面，要求用户更改用于生成绘制直方图的数据集的方程中的参数。 `

import PySimpleGUI as sg
import pandas as pd
import matplotlib.pyplot as plt
from matplotlib.backends.backend_tkagg import FigureCanvasTkAgg
from matplotlib.ticker import PercentFormatter


def read_table():
    sg.set_options(auto_size_buttons=True)
    filename = sg.popup_get_file(
        'Dataset to read',
        title='Dataset to read',
        no_window=True,
        file_types=(("CSV Files", ".csv"), ("Text Files", "*.txt")))
    # --- populate table with file contents --- #
    if filename == '':
        return

    colnames_prompt = sg.popup_yes_no('Does this file have column names already?')
    nan_prompt = sg.popup_yes_no('Drop NaN entries?')

    if filename is not None:
        fn = filename.split('/')[-1]
        try:
            if colnames_prompt == 'Yes':
                df = pd.read_csv(filename, sep=',', engine='python')
                # Uses the first row (which should be column names) as columns names
                header_list = list(df.columns)
                # Drops the first row in the table (otherwise the header names and the first row will be the same)
                data = df[1:].values.tolist()
            else:
                df = pd.read_csv(filename, sep=',', engine='python', header=None)
                # Creates columns names for each column ('column0', 'column1', etc)
                header_list = ['column' + str(x) for x in range(len(df.iloc[0]))]
                df.columns = header_list
                # read everything else into a list of rows
                data = df.values.tolist()
            # NaN drop?
            if nan_prompt == 'Yes':
                df = df.dropna()

            return df, data, header_list, fn
        except:
            sg.popup_error('Error reading file')
            return


def show_table(data, header_list, fn):
    layout = [
        [sg.Table(values=data,
                  headings=header_list,
                  font='Helvetica',
                  pad=(25, 25),
                  display_row_numbers=False,
                  auto_size_columns=True,
                  num_rows=min(25, len(data)))]
    ]

    window = sg.Window(fn, layout, grab_anywhere=False)
    event, values = window.read()
    window.close()


def show_stats(df):
    stats = df.describe().T
    header_list = list(stats.columns)
    data = stats.values.tolist()
    for i, d in enumerate(data):
        d.insert(0, list(stats.index)[i])
    header_list = ['Summary Statistics'] + header_list
    layout = [
        [sg.Table(values=data[12:],
                  headings=header_list,
                  font='Helvetica',
                  pad=(12, 12),
                  display_row_numbers=False,
                  auto_size_columns=True,
                  num_rows=min(25, len(data)))]
    ]

    window = sg.Window("Statistics", layout, grab_anywhere=False)
    event, values = window.read()
    window.close()


def plot_fig(df):
    """
    Plots
    """
    # Identify link to arrays
    input_file = r"input_csv_files/testfile.csv"

    # Read input files using pandas dataframe
    data_hss = pd.read_csv(input_file, na_values=':', usecols=['result1'])
    data_sp = pd.read_csv(input_file, na_values=':', usecols=['result2'])

    # Read input files from one file
    data = pd.read_csv(input_file, na_values=':', usecols=['result1', 'result2'])

    # Prepare files by eliminating NaN values
    dataDrop_hss = data_hss.dropna(how='any', subset=['result1'], axis=0)
    dataDrop_sp = data_sp.dropna(how='any', subset=['result2'], axis=0)

    # Make a separate list for each histogram
    hss = list(dataDrop_hss['result1'])
    s_p = list(dataDrop_sp['result2'])

    # Assign colors to each histogram and the legend names
    colors = ['tab:blue', 'tab:orange']
    names = ['hss', 'sp']

    # Plot the histogram
    fig, ax1 = plt.subplots()
    ax1.hist([hss, s_p], bins=20, label=names, color=colors)
    ax1.set_xlim(-0.14, 0.30)
    ax1.set_ylim(0, 120)
    plt.gca().xaxis.set_major_formatter(PercentFormatter(1))

    # Format the plots
    plt.xlabel('Return')
    plt.ylabel('Count')
    plt.title('hss and sp')
    plt.legend(loc='upper right')

    # ------------------------------- END OF YOUR MATPLOTLIB CODE -------------------------------

    # ------------------------------- Beginning of Matplotlib helper code -----------------------

    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

    # ------------------------------- Beginning of GUI CODE -------------------------------

    # Define the window layout
    layout = [[sg.Text('Plot test')],
              [sg.Canvas(key='-CANVAS-')],
              [sg.Button('Ok')]]

    # create the form and show it without the plot
    window = sg.Window('graph', layout, location=(0, 0), finalize=True, element_justification='center',
                       font='Helvetica 18', resizable=True)

    # Add the plot to the window
    draw_figure(window['-CANVAS-'].TKCanvas, fig)

    event, values = window.read()

    window.close()


def main():
    df, data, header_list, fn = read_table()

    # Show data?
    show_prompt = sg.popup_yes_no('Show the dataset?')
    if show_prompt == 'Yes':
        show_table(data, header_list, fn)

    # Show stats?
    stats_prompt = sg.popup_yes_no('Show the descriptive stats?')
    if stats_prompt == 'Yes':
        show_stats(df)

    # Show a plot?
    plot_prompt = sg.popup_yes_no('Show a histogram plot?')
    if plot_prompt == 'Yes':
        plot_fig(df)


# Executes main
if __name__ == '__main__':
    main()

`

这是我得到的汇总统计输出，但我没有用户界面供用户更改输入变量以操作图表和汇总统计

直方图：

汇总统计：

Answer 1

您需要花时间为您的应用程序设计 GUI，如果可能的话，一个主窗口。

下面为你展示示例

from pathlib import Path

import PySimpleGUI as sg
import pandas as pd
import matplotlib.pyplot as plt
from matplotlib.backends.backend_tkagg import FigureCanvasTkAgg
from matplotlib.ticker import PercentFormatter


def load_dataset(filename, headings, nan):

    if not Path(filename).is_file():
        return None
    window['Status'].update('file loading ...')
    window.refresh()

    try:
        if headings:
            df = pd.read_csv(filename, sep=',', engine='python')
            header_list = list(df.columns)
            data = df[1:].values.tolist()
        else:
            df = pd.read_csv(filename, sep=',', engine='python', header=None)
            header_list = ['column' + str(x) for x in range(len(df.iloc[0]))]
            df.columns = header_list
            data = df.values.tolist()
        if nan:
            df = df.dropna()
        window['Status'].update('File loaded')
        return df, data, header_list, Path(filename).name
    except:
        window['Status'].update('Error reading file')
        return None


def show_table(data, header_list, fn):

    layout = [
        [sg.Table(values=data,
                  headings=header_list,
                  font='Helvetica',
                  pad=(25, 25),
                  display_row_numbers=False,
                  auto_size_columns=True,
                  num_rows=min(25, len(data)))]
    ]

    sg.Window(fn, layout).read(close=True)

def show_statistics(df):

    stats = df.describe().T
    header_list = list(stats.columns)
    data = stats.values.tolist()
    for i, d in enumerate(data):
        d.insert(0, list(stats.index)[i])
    header_list = ['Summary Statistics'] + header_list

    layout = [
        [sg.Table(values=data[12:],
                  headings=header_list,
                  font='Helvetica',
                  pad=(12, 12),
                  display_row_numbers=False,
                  auto_size_columns=True,
                  num_rows=min(25, len(data)))]
    ]

    sg.Window("Statistics", layout).read(close=True)

def show_histogram(df):
    """
    Plots
    """
    # Identify link to arrays
    input_file = r"input_csv_files/testfile.csv"

    # Read input files using pandas dataframe
    data_hss = pd.read_csv(input_file, na_values=':', usecols=['result1'])
    data_sp = pd.read_csv(input_file, na_values=':', usecols=['result2'])

    # Read input files from one file
    data = pd.read_csv(input_file, na_values=':', usecols=['result1', 'result2'])

    # Prepare files by eliminating NaN values
    dataDrop_hss = data_hss.dropna(how='any', subset=['result1'], axis=0)
    dataDrop_sp = data_sp.dropna(how='any', subset=['result2'], axis=0)

    # Make a separate list for each histogram
    hss = list(dataDrop_hss['result1'])
    s_p = list(dataDrop_sp['result2'])

    # Assign colors to each histogram and the legend names
    colors = ['tab:blue', 'tab:orange']
    names = ['hss', 'sp']

    # Plot the histogram
    fig, ax1 = plt.subplots()
    ax1.hist([hss, s_p], bins=20, label=names, color=colors)
    ax1.set_xlim(-0.14, 0.30)
    ax1.set_ylim(0, 120)
    plt.gca().xaxis.set_major_formatter(PercentFormatter(1))

    # Format the plots
    plt.xlabel('Return')
    plt.ylabel('Count')
    plt.title('hss and sp')
    plt.legend(loc='upper right')

    # ------------------------------- END OF YOUR MATPLOTLIB CODE -------------------------------

    # ------------------------------- Beginning of Matplotlib helper code -----------------------

    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

    # ------------------------------- Beginning of GUI CODE -------------------------------

    # Define the window layout
    layout = [[sg.Text('Plot test')],
              [sg.Canvas(key='-CANVAS-')],
              [sg.Button('Ok')]]

    # create the form and show it without the plot
    window = sg.Window('graph', layout, location=(0, 0), finalize=True, element_justification='center',
                       font='Helvetica 18', resizable=True)

    # Add the plot to the window
    draw_figure(window['-CANVAS-'].TKCanvas, fig)

    window.read(close=True)


layout = [
    [sg.Text("Dateset"),
     sg.Checkbox("With column names", enable_events=True, key="Headings"),
     sg.Checkbox("Drop NaN entries", enable_events=True, key="NaN")],
    [sg.Input(key='Dataset'), sg.FileBrowse()],
    [sg.Button("Show Dataset"), sg.Button("Show Statistics"), sg.Button("Show Histogram")],
    [sg.StatusBar('', size=20, expand_x=True, key='Status')],
]

window = sg.Window("Dataset", layout, finalize=True)
filename = None
load = False
while True:

    event, values = window.read()

    if event == sg.WINDOW_CLOSED:
        break
    if filename != values["Dataset"] or event in ("Headings", "NaN"):
        filename, headings, nan = values["Dataset"], values['Headings'], values['NaN']
        result = load_dataset(filename, headings, nan)
        if result:
            load = True
            df, data, header_list, fn = result
        else:
            load = False
    if not load:
        continue
    if event == "Show Dataset":
        show_table(data, header_list, fn)
    elif event == "Show Statistics":
        show_statistics(df)
    elif event == "Show Histogram":
        show_histogram(df)

window.close()