Matplotlib分组条形图首条不显示及Y轴调整技术咨询

问题描述 #

使用Matplotlib绘制分组条形图时，第一个条形完全未显示。推测是Y轴刻度设置问题，尝试多种方法未解决。当前Y轴从最小数值开始，需在不破坏原有刻度比例的前提下修复。

原代码 #

from matplotlib import pyplot as plt
import numpy as np

x = np.arange(4)  # the label locations
width = 0.2  # the width of the bars

plt.bar(x-0.2, one_digit_10_pre_run, width, color='cyan')
plt.bar(x, two_digit_10_pre_run, width, color='orange')
plt.bar(x+0.2, three_digit_10_pre_run,width,color='green')

plt.xticks(x, ['10', '20', '30', "100"])
plt.xlabel("Amount of Input Values")
plt.ylabel("Execution Times [Seconds]")
plt.legend(["0 Digits extra", "1 Digit Extra", "2 Digit Extra"])
plt.show()

原始数据 #

['0.021636664867401074', '0.023439047336578315', '0.023759815692901566', 027013454437255814]
['  0.021725461034491467', '  0.020219009701568258', '  0.021416027545928907', '  0.024219112396240184']
['  0.02276699304580684', '  0.021797661781310988', '  0.025862150192260697', '  0.026579418182373003']

问题分析与修复步骤 #

1. 数据类型修正 #

原始数据存在两个问题：

• 第一个列表最后一个元素027013454437255814是非法八进制整数（以0开头），与其他字符串格式的浮点数不统一；
• 部分数据带有多余空格。

先将所有数据转换为浮点数：

one_digit_10_pre_run = [float(num.strip()) for num in ['0.021636664867401074', '0.023439047336578315', '0.023759815692901566', '0.027013454437255814']]
two_digit_10_pre_run = [float(num.strip()) for num in ['  0.021725461034491467', '  0.020219009701568258', '  0.021416027545928907', '  0.024219112396240184']]
three_digit_10_pre_run = [float(num.strip()) for num in ['  0.02276699304580684', '  0.021797661781310988', '  0.025862150192260697', '  0.026579418182373003']]

2. Y轴刻度调整 #

Y轴从最小值开始时，条形底部会被坐标轴截断。通过设置略低于最小值的Y轴下限，既能显示完整条形，又不破坏原有刻度比例：

# 获取所有数据的最小值
all_data = one_digit_10_pre_run + two_digit_10_pre_run + three_digit_10_pre_run
min_val = min(all_data)
# 设置Y轴下限为最小值的95%
plt.ylim(bottom=min_val * 0.95)

3. 完整修正代码 #

from matplotlib import pyplot as plt
import numpy as np

# 修正后的数据
one_digit_10_pre_run = [float(num.strip()) for num in ['0.021636664867401074', '0.023439047336578315', '0.023759815692901566', '0.027013454437255814']]
two_digit_10_pre_run = [float(num.strip()) for num in ['  0.021725461034491467', '  0.020219009701568258', '  0.021416027545928907', '  0.024219112396240184']]
three_digit_10_pre_run = [float(num.strip()) for num in ['  0.02276699304580684', '  0.021797661781310988', '  0.025862150192260697', '  0.026579418182373003']]

x = np.arange(4)
width = 0.2

plt.bar(x-0.2, one_digit_10_pre_run, width, color='cyan')
plt.bar(x, two_digit_10_pre_run, width, color='orange')
plt.bar(x+0.2, three_digit_10_pre_run, width, color='green')

plt.xticks(x, ['10', '20', '30', "100"])
plt.xlabel("输入值数量")
plt.ylabel("执行时间 [秒]")
plt.legend(["0位额外数字", "1位额外数字", "2位额外数字"])

# 调整Y轴下限
all_data = one_digit_10_pre_run + two_digit_10_pre_run + three_digit_10_pre_run
min_val = min(all_data)
plt.ylim(bottom=min_val * 0.95)

plt.show()

内容的提问来源于stack exchange，提问作者tapir36