Как построить горизонтальную столбчатую диаграмму с положительным или отрицательным процентным значением со значениями, отображаемыми с помощью Python

Question

У меня есть фрейм данных, но я пытаюсь построить график, и мне трудно. Ребята, не могли бы вы помочь мне с этим? Я пытался Google, но я не могу найти действительный ресурс.

Answer 1

Друг, твой график попался на глаза, и я решил его создать. Я многому научился в процессе, и я надеюсь, что это может быть полезно для вас:

Входные данные:

    Completely Dissatisfied Somewhat Dissatisfied   Dissatisfied    Neither Satisfied nor Dissatisfied  Somewhat Satisfied  Satisfied   Completely Satisfied
Q1  10  10  10  10  10  20  30
Q2  0   0   20  20  30  20  10
Q3  10  20  0   30  20  20  20
Q4  0   0   10  10  30  20  10

Код изображения:

import matplotlib.pyplot as plt
import matplotlib.patches as patches
import numpy as np
import pandas as pd

# Read in data
df = pd.read_csv('plot_data.csv', index_col=0)
df = df/100 # Convert to percentages

def custom_plot(df, example=False):

    def plot_rect(bottom, left, width, height, color = 'C0'):
        ax.add_patch(patches.Rectangle(
                (left, bottom), width, height, linewidth=1, edgecolor=color, facecolor=color))

    # Create figure and axes
    fig, ax = plt.subplots(1)

    # Define axis ticks ticks
    plt.xticks(np.arange(-1,1.25,0.25), np.arange(-100,125,25))
    plt.yticks(np.arange(0,1.2,0.2), np.arange(0,1.2,0.2))

    # Define axis limits
    plt.ylim(0.05,0.95)
    plt.xlim(-1.125, 1.125)

    # Move gridlines to the back of the plot
    plt.gca().set_axisbelow(True)

    # Change color of plot edges
    ax.spines['left'].set_color('lightgray')
    ax.spines['right'].set_color('lightgray')
    ax.spines['top'].set_color('lightgray')

    # Hide y axis ticks
    plt.gca().tick_params(axis='y', colors='w')

    # Turn on gridlines and set color
    plt.grid(b=True, axis='both', color='lightgray', alpha=0.5, linewidth=1.5)

    # Add lines
    plt.axvline(x=0, c='lightgray')
    plt.axhline(y=0.5, c='black')

    # Add x label
    plt.xlabel('Percent', fontsize=14)

    # Define color scheme from negative to positive
    colors = ['firebrick', 'sandybrown', 'navajowhite', 
              'khaki', 'lightcyan', 'skyblue', 'steelblue']

    # Process data to plot
    try:
        array = [df.iloc[0,:].values, 
                 df.iloc[1,:].values, 
                 df.iloc[2,:].values, 
                 df.iloc[3,:].values]
    except:
        print('Plotting example data')
        example = True

    if example == True:
        # Example data
        array = [np.array([0.05, 0.1, 0.2, 0.2, 0.3, 0.1, 0.05]),
                 np.array([0, 0.1, 0.1, 0.3, 0.2, 0.2, 0.1]),
                 np.array([0.1, 0.2, 0.2, 0.3, 0.1, 0.05, 0.05]),
                 np.array([0.2, 0.1, 0.2, 0.2, 0.1, 0.1, 0.1])]

        # Example data column names
        df = pd.DataFrame(columns=['Completely Dissatisfied',
                                   'Somewhat Dissatisfied',
                                   'Dissatisfied',
                                   'Neither Satisfied nor Dissatisfied',
                                   'Somewhat Satisfied',
                                   'Satisfied',
                                   'Completely Satisfied'])
    # Compute average statistics
    hi = [sum(x[4:]) for x in array]
    med = [x[3] for x in array]
    lo = [sum(x[0:3]) for x in array]

    # Define function to process input data into rectangle left corner indices
    def process_data(array):
        left = np.zeros_like(array)
        mid = array[3]/2
        left[0] = -np.sum(array[0:3]) - mid
        left[1] = -np.sum(array[1:3]) - mid
        left[2] = -np.sum(array[2:3]) - mid
        left[3] = -mid
        left[4] = mid
        left[5] = np.sum(array[4:5]) + mid
        left[6] = np.sum(array[4:6]) + mid
        width = array
        return left, width

    left = {}
    width = {}
    for i in range(4):
        left[i], width[i] = process_data(array[i])

    # Plot boxes
    height = 0.13
    bottom = 0.135
    for i in range(len(array)):
        for j in range(len(array[i])):
            plot_rect(bottom=bottom+i*0.2, left=left[i][j], width=width[i][j], height=height, color = colors[j])

    # Plot category labels
    plt.text(-1.1,0.9,'Unfavorable', style='italic', 
             horizontalalignment='left', verticalalignment='center')
    plt.text(0,0.9,'Neutral', style='italic', 
             horizontalalignment='center', verticalalignment='center')
    plt.text(1.1,0.9,'Favorable', style='italic', 
             horizontalalignment='right', verticalalignment='center')

    # Plot percentages
    for i in range(len(med)):
        plt.text(-1,0.2*(i+1),'{0:.0%}'.format(lo[i]), 
                 horizontalalignment='left', verticalalignment='center')
        plt.text(0,0.2*(i+1),'{0:.0%}'.format(med[i]), 
                 horizontalalignment='center', verticalalignment='center')
        plt.text(1,0.2*(i+1),'{0:.0%}'.format(hi[i]), 
                 horizontalalignment='right', verticalalignment='center')

    # Create legend
    fig, ax = plt.subplots(1, figsize=(6,2))
    plt.axis('off')
    plt.gca().set_aspect('equal', adjustable='box')

    # Plot colored circles
    legend_left = [-0.9, -0.6, -0.3, 0, 0.30, 0.6, 0.9]
    for i in range(len(colors)):
        plot_rect(bottom=0, left=legend_left[i], width=0.2, height=0.2, color = colors[i])

    # Plot labels 1-6
    for i in range(0,6,2):
        plt.text(-0.8+0.3*i, 0.25, df.columns[i].replace(' ', '\n'), 
                 horizontalalignment='center', verticalalignment='bottom')
        plt.text(-0.5+0.3*i, -0.05, df.columns[i+1].replace(' ', '\n'), 
                 horizontalalignment='center', verticalalignment='top')

    # Plot last label
    plt.text(1, 0.25, df.columns[6].replace(' ', '\n'), 
             horizontalalignment='center', verticalalignment='bottom')

    # Plot label title
    plt.text(-1, 0.1, 'Scale', fontsize=14,
             horizontalalignment='right', verticalalignment='center')

    plt.gca().autoscale(enable=True, axis='both', tight=None)

#custom_plot('example')
custom_plot(df)