AnnotationBbox与errorbar同绘时不显示的原因及解决咨询

我用以下Python代码实现带误差棒的Matplotlib绘图，并在图旁添加AnnotationBbox文本框：

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.offsetbox import AnnotationBbox, TextArea

fig, ax = plt.subplots(layout="constrained")

f = np.array([2.15, 2.22, 2.3, 2.43, 2.54, 2.7, 2.79, 2.91, 3.04, 3.2, 3.29])
f_err = 0.01
d = np.array(
    [
        0.3393,
        0.3458,
        0.3622666666666667,
        0.38306666666666667,
        0.4004,
        0.429,
        0.43939999999999996,
        0.45239999999999997,
        0.481,
        0.5044,
        0.5174,
    ]
)
d_err = 0.0052

ax.errorbar(f, d, xerr=f_err, yerr=d_err, fmt=".", label="Measurements")

text_area = TextArea("Hello there!!!!!!")
text_box = AnnotationBbox(
    text_area, (1, 1), box_alignment=(0, 1), boxcoords="axes fraction"
)
ax.add_artist(text_box)

ax.legend()
plt.show()

此前代码运行正常，但现在运行时文本框的空间被预留（修改文本长度会改变该空间），但文本内容本身不显示；注释掉ax.errorbar相关代码后，文本框可正常显示。

问题原因 #

这是constrained_layout布局模式与AnnotationBbox的交互异常导致的：当使用layout="constrained"时，errorbar元素会触发布局的重新计算，使得TextArea的文本渲染被抑制，仅保留了布局预留空间但不显示实际内容。

解决方法 #

方法一：更换布局模式 #

将constrained_layout替换为tight_layout，修改初始化代码并在绘图最后手动调用布局调整：

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.offsetbox import AnnotationBbox, TextArea

fig, ax = plt.subplots()  # 移除layout="constrained"

f = np.array([2.15, 2.22, 2.3, 2.43, 2.54, 2.7, 2.79, 2.91, 3.04, 3.2, 3.29])
f_err = 0.01
d = np.array(
    [
        0.3393,
        0.3458,
        0.3622666666666667,
        0.38306666666666667,
        0.4004,
        0.429,
        0.43939999999999996,
        0.45239999999999997,
        0.481,
        0.5044,
        0.5174,
    ]
)
d_err = 0.0052

ax.errorbar(f, d, xerr=f_err, yerr=d_err, fmt=".", label="Measurements")

text_area = TextArea("Hello there!!!!!!")
text_box = AnnotationBbox(
    text_area, (1, 1), box_alignment=(0, 1), boxcoords="axes fraction"
)
ax.add_artist(text_box)

ax.legend()
plt.tight_layout()  # 添加手动布局调整
plt.show()

方法二：保留constrained_layout并调整AnnotationBbox参数 #

添加annotation_clip=False避免文本被裁剪，同时显式设置文本框背景（防止与画布背景融合）：

import numpy as np
import matplotlib.pyplot as plt
from matplotlib.offsetbox import AnnotationBbox, TextArea

fig, ax = plt.subplots(layout="constrained")

f = np.array([2.15, 2.22, 2.3, 2.43, 2.54, 2.7, 2.79, 2.91, 3.04, 3.2, 3.29])
f_err = 0.01
d = np.array(
    [
        0.3393,
        0.3458,
        0.3622666666666667,
        0.38306666666666667,
        0.4004,
        0.429,
        0.43939999999999996,
        0.45239999999999997,
        0.481,
        0.5044,
        0.5174,
    ]
)
d_err = 0.0052

ax.errorbar(f, d, xerr=f_err, yerr=d_err, fmt=".", label="Measurements")

text_area = TextArea("Hello there!!!!!!")
text_box = AnnotationBbox(
    text_area, (1, 1), box_alignment=(0, 1), boxcoords="axes fraction",
    annotation_clip=False,  # 禁止裁剪注释
    bboxprops={"facecolor": "white", "edgecolor": "black"}  # 设置文本框背景和边框
)
ax.add_artist(text_box)

ax.legend()
plt.show()

内容的提问来源于stack exchange，提问作者Shai Avr