如何在带有日期x轴的绘图上绘制 FancyBboxPatch。FancyBboxPatch 应该在特定的时间段内延伸，类似于甘特。

import matplotlib.pyplot as plt
import pandas as pd

data = [('2019-04-15', 'Start'), ('2019-05-01', 'Stop')]
date, labels = zip(*data)
frame = pd.DataFrame({'bla': labels}, index=pd.DatetimeIndex(date))

fig, ax = plt.subplots(figsize=(12.5, 2.5))

for index, row in frame.iterrows():

    ax.text(index, 0.5, row['bla'])

ax.set_xlim([frame.index[0], frame.index[-1]])

plt.show()

我试着从Matplotlib网站复制粘贴和调整代码。但是，这导致了与日期轴和补丁宽度有关的错误。

更新

我尝试了下面的代码，但是它返回TypeError: unsupported operand type(s) for -: 'Timestamp' and 'float'。

import matplotlib.pyplot as plt
import pandas as pd
import matplotlib.patches as mpatches

data = [('2019-04-15', 'Start'), ('2019-05-01', 'Stop')]
date, labels = zip(*data)
frame = pd.DataFrame({'bla': labels}, index=pd.DatetimeIndex(date))

fig, ax = plt.subplots(figsize=(12.5, 2.5))

for index, row in frame.iterrows():

    ax.text(index, 0.5, row['bla'])

ax.set_xlim([frame.index[0], frame.index[-1]])

ax.add_patch(
    mpatches.FancyBboxPatch(
        (frame.index[0],0.5),  # xy
        100,  # width
        0.5,  # heigth
        boxstyle=mpatches.BoxStyle("Round", pad=0.02)
        )
    )

plt.show()