PyQt6与Matplotlib集成报错：无法导入backend_qt6agg的问题咨询

问题背景 #

开发基于PyQt6和Matplotlib的硬件监控应用时，运行程序出现模块找不到的错误。

错误信息 #

运行程序时报错：

Traceback (most recent call last):
  File "d:\programforpyqt6\Program\program.py", line 4, in <module>
    from matplotlib.backends.backend_qt6agg import FigureCanvasQTAgg as FigureCanvas
ModuleNotFoundError: No module named 'matplotlib.backends.backend_qt6agg'

代码示例 #

from PyQt6.QtWidgets import QApplication, QMainWindow, QVBoxLayout, QWidget, QPushButton
import sys
from matplotlib.backends.backend_qt6agg import FigureCanvasQTAgg as FigureCanvas
import matplotlib.pyplot as plt
from matplotlib.figure import Figure
import platform
import psutil


class HardwareMonitorApp(QMainWindow):
    def __init__(self):
        super().__init__()

        self.central_widget = QWidget(self)
        self.setCentralWidget(self.central_widget)

        self.layout = QVBoxLayout(self.central_widget)

        self.cpu_button = QPushButton('Show CPU Usage', self)
        self.cpu_button.clicked.connect(self.show_cpu_usage)
        self.layout.addWidget(self.cpu_button)

        self.memory_button = QPushButton('Show Memory Usage', self)
        self.memory_button.clicked.connect(self.show_memory_usage)
        self.layout.addWidget(self.memory_button)

        self.setWindowTitle('Hardware Monitor App')

    def show_cpu_usage(self):
        x = list(range(1, 11))
        y = psutil.cpu_percent(interval=1, percpu=True)

        self.plot_graph(x, y, 'CPU Usage', 'Core', 'Usage (%)')

    def show_memory_usage(self):
        virtual_memory = psutil.virtual_memory()
        swap_memory = psutil.swap_memory()

        labels = ['Used', 'Available', 'Total']
        sizes = [virtual_memory.used, virtual_memory.available, virtual_memory.total]

        self.plot_pie_chart(labels, sizes, 'Memory Usage')

    def plot_graph(self, x, y, title, xlabel, ylabel):
        figure = Figure(figsize=(6, 4), tight_layout=True)
        plot = figure.add_subplot(1, 1, 1)
        plot.plot(x, y, marker='o')
        plot.set_title(title)
        plot.set_xlabel(xlabel)
        plot.set_ylabel(ylabel)

        canvas = FigureCanvas(figure)
        self.layout.addWidget(canvas)

    def plot_pie_chart(self, labels, sizes, title):
        figure, ax = plt.subplots()
        ax.pie(sizes, labels=labels, autopct='%1.1f%%', startangle=90)
        ax.axis('equal')  # Equal aspect ratio ensures that pie is drawn as a circle.
        ax.set_title(title)

        canvas = FigureCanvas(figure)
        self.layout.addWidget(canvas)


def main():
    app = QApplication(sys.argv)
    main_win = HardwareMonitorApp()
    main_win.show()
    sys.exit(app.exec())


if __name__ == '__main__':
    main()

补充环境信息 #

• Python版本：3.12.1
• 操作系统：Windows 10
• PyQt6版本：6.6.1
• Matplotlib版本：3.8.2
• PyQt6安装方式：pip install PyQt6
• Matplotlib安装方式：pip install matplotlib
• 虚拟环境：未使用
• 开发场景：PyQt6整合Matplotlib绘图的硬件监控应用

已尝试的解决步骤 #

• 执行pip show PyQt6确认PyQt6版本
• 卸载并重装Matplotlib：pip uninstall matplotlib + pip install matplotlib

疑问解答 #

1. PyQt6 6.6.1与Matplotlib 3.8.2是否存在兼容性问题？ #

不存在直接的兼容性问题。Matplotlib 3.8.x正式支持PyQt6，backend_qt6agg模块是Matplotlib针对PyQt6的官方后端实现，理论上两者版本匹配可以正常工作。

2. 是否有适配PyQt6与Matplotlib的其他导入语句？ #

有两种替代方案：

• 兼容导入（同时支持PyQt5和PyQt6）：

  try:
      from matplotlib.backends.backend_qt6agg import FigureCanvasQTAgg as FigureCanvas
  except ImportError:
      from matplotlib.backends.backend_qt5agg import FigureCanvasQTAgg as FigureCanvas
  

• 统一Qt后端导入（Matplotlib 3.7+支持）：

  from matplotlib.backends.backend_qtagg import FigureCanvasQTAgg as FigureCanvas
  

  该模块是Matplotlib引入的统一Qt后端，无需区分PyQt5/6版本，自动适配当前环境。

3. 该问题是否有已知问题或解决办法？ #

这个错误的常见原因及对应解决办法：

• Matplotlib未安装Qt相关依赖：执行pip install matplotlib[qt]，该命令会自动安装Matplotlib运行Qt后端所需的所有依赖包。
• Python环境混乱：创建并激活虚拟环境，重新安装依赖：

  python -m venv venv
  venv\Scripts\activate  # Windows系统
  pip install PyQt6 matplotlib[qt]
  

• Matplotlib安装不完整：彻底卸载后清理残留文件，再重新安装：

  pip uninstall -y matplotlib
  # 手动删除Python安装目录Lib\site-packages下的matplotlib残留文件夹
  pip install matplotlib[qt]
  

• Python版本适配问题：Python 3.12属于较新版本，部分库可能存在适配延迟，可尝试降级到Python 3.11后重新安装依赖。

内容的提问来源于stack exchange，提问作者RatRatARMY