堆积条形图的格式索引

Question

Hpw 我可以在几分钟和几小时内设置一个有用的索引吗？

csv_file = dir_path + "/stacktest.csv"
with open(csv_file, newline='') as csv_file:
    data = pd.read_csv(csv_file, sep=',')
df = pd.DataFrame(data)
df = df[['seconds', 'marker', 'data1', 'data2', 'data3']]
df['seconds'] = df['seconds'].astype(str)
df = df.set_index('seconds')
dfStacked = df[['data1', 'data2']]

ax = dfStacked.plot(kind='bar', stacked=True, alpha=0.5)
import matplotlib.dates as mdates
majorFmt = mdates.DateFormatter('%H:%M')
ax.xaxis.set_major_locator(mdates.MinuteLocator(interval=30))
ax.xaxis.set_major_formatter(majorFmt)
plt.plot(df.index, df['data3'], linestyle='solid', color='blue', alpha=0.4, label='data1')
plt.show()

如果我删除 DateFormatter，索引似乎有问题。

通过将行更改为：

#majorFmt = mdates.DateFormatter('%H:%M')
ax.xaxis.set_major_locator(mdates.MinuteLocator(interval=60*60))
#ax.xaxis.set_major_formatter(majorFmt)

有一个 [121,377,...] 的 x 索引 121是秒值，它在2分钟处设置一个标记，间隔为60*60。

示例数据

seconds,marker,data1,data2,data3,data4
0,B,0,0,0,0
59,C,42000,8000,369000,0
74,B,42000,8000,369000,283041
121,B,42000,8000,369000,283041
179,B,42000,8000,369000,283041
239,B,42000,8000,369000,283041
304,B,42000,8000,369000,283041
360,B,42000,8000,369000,283041
377,A,42000,8000,369000,283041
420,B,42000,8000,369000,283041
493,B,42000,8000,369000,283041
540,B,42000,8000,369000,283041
600,B,42000,8000,369000,283041
659,B,42000,8000,369000,283041
719,B,64000,8000,412000,283041
780,B,64000,8000,412000,283041
840,B,64000,8000,412000,283041
880,A,64000,8000,412000,283041
900,B,64000,8000,412000,283041
961,B,64000,8000,412000,283041
1020,B,64000,8000,412000,283041
1079,B,64000,8000,412000,283041
1141,B,64000,8000,412000,283041
1200,B,64000,8000,412000,283041
1260,B,64000,8000,412000,283041
1320,B,64000,8000,412000,283041
1365,A,64000,8000,412000,283041
1382,B,64000,8000,412000,283041
1440,B,64000,8000,412000,283041
1498,B,64000,8000,412000,283041
1559,B,64000,8000,412000,283041
1621,B,64000,8000,412000,283041
1679,B,64000,8000,412000,283041
1740,B,64000,8000,412000,283041
1800,B,42000,8000,369000,283041
1830,A,42000,8000,369000,283041
1867,B,42000,8000,369000,283041
1921,B,42000,8000,369000,283041
1979,B,42000,8000,369000,283041
2040,B,42000,8000,369000,283041
2099,B,42000,8000,369000,283041
2159,B,42000,8000,369000,283041
2220,B,42000,8000,369000,283041
2272,A,42000,8000,369000,283041
2288,B,42000,8000,369000,283041
2341,B,42000,8000,369000,283041
2400,B,42000,8000,369000,283041
2460,B,42000,8000,369000,283041
2520,B,42000,8000,369000,283041
2579,B,42000,8000,369000,283041
2640,B,42000,8000,369000,283041
2700,B,42000,8000,369000,283041
2720,A,42000,8000,369000,283041
2759,B,42000,8000,369000,283041
2833,B,28000,14000,248000,260096
2880,B,28000,14000,248000,247808
2940,B,14000,28000,124000,123904
3000,B,0,42000,0,0
3060,B,0,42000,0,0
3120,B,0,42000,0,0
3136,A,0,42000,0,0
3180,B,0,42000,0,0
3251,B,0,42000,0,0
3267,D,0,42000,0,0
3300,B,0,42000,0,0
3359,B,0,42000,0,0
3419,B,0,42000,0,0

Answer 1

您可以编写自定义格式化程序以将秒数显示为小时和分钟。

要用数字 x-axis 绘制条形图，可以使用 matplotlib 的 bar()。条形的宽度各不相同，它们可以根据连续秒数之间的差异来计算。下面的代码显示了相互粘附的条形图。设置边缘颜色（ax.bar(..., ec='white', lw=1) 会显示一个小的分离。

import matplotlib.pyplot as plt
from matplotlib.ticker import MultipleLocator
import pandas as pd
import numpy as np

def hms_formatter(x, pos):
    seconds = int(x)
    minutes = seconds // 60
    seconds %= 60
    hours = minutes // 60
    minutes %= 60
    if hours == 0:
        return f'{minutes:2d}:{seconds:02d}:'
    else:
        return f'{hours:2d}:{minutes:02d}:{seconds:02d}:'

df = pd.read_csv(...)
fig, ax = plt.subplots(figsize=(15, 5))
bottom = 0
widths = np.diff(df['seconds'])
widths = np.append(widths, widths[-1])
for col in ['data1', 'data2', 'data3']:
    ax.bar(df['seconds'], df[col], bottom=bottom, width=widths,
           align='edge', label=col)
    bottom += df[col]

ax.plot(df['seconds'], df['data3'], linestyle='solid', color='crimson', lw=3, alpha=0.4, label='data3 (unstacked)')

ax.margins(x=0.01)
ax.xaxis.set_major_locator(MultipleLocator(10 * 60))
ax.xaxis.set_major_formatter(hms_formatter)
ax.legend()
plt.tight_layout()
plt.show()

堆积条形图的格式索引

format index at stacked bar plot

plot

matplotlib

bar-chart

dataframe

stacked-chart