Matplotlib子图报错：'Axes'对象无'is_first_col'属性

问题重现 #

运行以下代码时触发AttributeError：

import matplotlib.pyplot as plt
fig, ax = plt.subplots(1,2)
import pandas as pd
df = pd.DataFrame([1,2,3])
df.plot(ax=ax[0])

环境版本 #

• matplotlib: '3.8.4'
• pandas: '1.2.0'
• python: 3.11.0

报错信息 #

---------------------------------------------------------------------------
AttributeError                            Traceback (most recent call last)
Cell In[3], line 5
      3 import pandas as pd
      4 df = pd.DataFrame([1,2,3])
----> 5 df.plot(ax=ax[0])

File ~/miniforge3/envs/ExMAS/lib/python3.11/site-packages/pandas/plotting/_core.py:955, in PlotAccessor.__call__(self, *args, **kwargs)
    952             label_name = label_kw or data.columns
    953             data.columns = label_name
---> 955 return plot_backend.plot(data, kind=kind, **kwargs)

File ~/miniforge3/envs/ExMAS/lib/python3.11/site-packages/pandas/plotting/_matplotlib/__init__.py:61, in plot(data, kind, **kwargs)
     59         kwargs["ax"] = getattr(ax, "left_ax", ax)
     60 plot_obj = PLOT_CLASSES[kind](data, **kwargs)
---> 61 plot_obj.generate()
     62 plot_obj.draw()
     63 return plot_obj.result

File ~/miniforge3/envs/ExMAS/lib/python3.11/site-packages/pandas/plotting/_matplotlib/core.py:283, in MPLPlot.generate(self)
    281 self._add_table()
    282 self._make_legend()
---> 283 self._adorn_subplots()
    285 for ax in self.axes:
    286     self._post_plot_logic_common(ax, self.data)

File ~/miniforge3/envs/ExMAS/lib/python3.11/site-packages/pandas/plotting/_matplotlib/core.py:485, in MPLPlot._adorn_subplots(self)
    483     all_axes = self._get_subplots()
    484     nrows, ncols = self._get_axes_layout()
---> 485     handle_shared_axes(
    486         axarr=all_axes,
    487         nplots=len(all_axes),
    488         naxes=nrows * ncols,
    489         nrows=nrows,
    490         ncols=ncols,
    491         sharex=self.sharex,
    492         sharey=self.sharey,
    493     )
    495 for ax in self.axes:
    496     if self.yticks is not None:

File ~/miniforge3/envs/ExMAS/lib/python3.11/site-packages/pandas/plotting/_matplotlib/tools.py:400, in handle_shared_axes(axarr, nplots, naxes, nrows, ncols, sharex, sharey)
    395 if ncols > 1:
    396     for ax in axarr:
    397         # only the first column should get y labels -> set all other to
    398         # off as we only have labels in the first column and we always
    399         # have a subplot there, we can skip the layout test
---> 400         if ax.is_first_col():
    401             continue
    402         if sharey or _has_externally_shared_axis(ax, "y"):

AttributeError: 'Axes' object has no attribute 'is_first_col'

问题原因 #

pandas 1.2.0的绘图代码依赖Matplotlib Axes对象的is_first_col()方法，但该方法在Matplotlib 3.8.4中已被移除，属于版本兼容冲突。

解决方案 #

方案1：升级pandas版本 #

将pandas升级到与Matplotlib 3.8.4兼容的版本（建议升级到1.5.x及以上，或最新稳定版），执行命令：

pip install --upgrade pandas

方案2：降级Matplotlib版本 #

将Matplotlib降级到pandas 1.2.0支持的版本（建议降级到3.3.x及以下），执行命令：

pip install matplotlib==3.3.4

方案3：临时绕过（不推荐生产环境使用） #

手动给Axes对象添加is_first_col()方法，在调用df.plot()前插入以下代码：

def mock_is_first_col(self):
    return self.get_subplotspec().is_first_col()

import matplotlib.axes
matplotlib.axes.Axes.is_first_col = mock_is_first_col

内容的提问来源于stack exchange，提问作者Intelligent-Infrastructure