Python中Graphviz库render函数无法运行问题求助

我用Python的Graphviz库写了数据科学流程的代码，但render函数报错，注释掉这行代码能执行但没有任何输出。我需要生成PNG或PDF格式的流程图，也可以用其他库替代。

我的代码：

from graphviz import Graph 

# 实例化Graph对象
dot = Graph('Data Science Process', format='png')

# 添加节点
dot.node('A', 'Get Data')
dot.node('B', 'Clean, Prepare, & Manipulate Data')
dot.node('C', 'Train Model')
dot.node('D', 'Test Data')
dot.node('E', 'Improve')

# 连接节点
dot.edges(['AB', 'BC', 'CD', 'DE'])

# 保存图表（注释掉才能运行）
#dot.render('data_science_flowchart', view=True)

报错信息：

$ py graph2.py
py graph2.py
Traceback (most recent call last):
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\site-packages\graphviz\backend\execute.py", line 81, in run_check
    proc = subprocess.run(cmd, **kwargs)
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\subprocess.py", line 501, in run
    with Popen(*popenargs, **kwargs) as process:
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\subprocess.py", line 966, in __init__
    self._execute_child(args, executable, preexec_fn, close_fds,
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\subprocess.py", line 1435, in _execute_child
    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
FileNotFoundError: [WinError 2] 系统找不到指定的文件。

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "C:\Users\vince\graph2.py", line 17, in <module>
    dot.render('data_science_flowchart', view=True)
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\site-packages\graphviz\_tools.py", line 171, in wrapper
    return func(*args, **kwargs)
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\site-packages\graphviz\rendering.py", line 122, in render
    rendered = self._render(*args, **kwargs)
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\site-packages\graphviz\_tools.py", line 171, in wrapper
    return func(*args, **kwargs)
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\site-packages\graphviz\backend\rendering.py", line 324, in render
    execute.run_check(cmd,
  File "C:\Users\vince\AppData\Local\Programs\Python\Python310\lib\site-packages\graphviz\backend\execute.py", line 84, in run_check
    raise ExecutableNotFound(cmd) from e
graphviz.backend.execute.ExecutableNotFound: failed to execute WindowsPath('dot'), make sure the Graphviz executables are on your systems' PATH

一、修复Graphviz问题 #

报错核心是系统找不到Graphviz的dot可执行文件——你只装了Python的graphviz库，没装Graphviz本体，或者没把本体路径加到系统环境变量里。

方法1：安装Graphviz本体并配置环境变量 #

1. 下载对应Windows版本的Graphviz安装包，安装时勾选“Add Graphviz to the system PATH for all users”或“Current user”。
2. 安装完成后重启命令行或IDE，再运行代码即可生成图片。

方法2：直接在代码里指定dot路径（不用改环境变量） #

如果不想调整系统环境变量，直接在代码里添加dot.exe的路径（替换成你的实际安装路径）：

from graphviz import Graph 

dot = Graph('Data Science Process', format='png')
# 指定dot.exe的路径，注意用raw字符串或者转义斜杠
dot.engine = r'C:\Program Files\Graphviz\bin\dot.exe'

dot.node('A', 'Get Data')
dot.node('B', 'Clean, Prepare, & Manipulate Data')
dot.node('C', 'Train Model')
dot.node('D', 'Test Data')
dot.node('E', 'Improve')

dot.edges(['AB', 'BC', 'CD', 'DE'])

dot.render('data_science_flowchart', view=True)

二、用其他库快速实现 #

如果不想折腾Graphviz，试试下面几个新手友好的替代方案：

方案1：pyflowchart #

安装命令：pip install pyflowchart
代码示例：

from pyflowchart import Flowchart

# 定义流程步骤
steps = [
    "Get Data",
    "Clean, Prepare, & Manipulate Data",
    "Train Model",
    "Test Data",
    "Improve"
]

# 创建流程图
fc = Flowchart(steps[0])
current_node = fc.start_node
for step in steps[1:]:
    current_node = current_node.add_operation(step)
current_node.add_end()

# 导出为PNG（第一次运行会自动下载依赖）
fc.export("data_science_flowchart.png", format="png")

方案2：networkx + matplotlib #

安装命令：pip install networkx matplotlib
代码示例：

import networkx as nx
import matplotlib.pyplot as plt

# 创建有向图
G = nx.DiGraph()

# 添加节点和标签
node_labels = {
    'A': 'Get Data',
    'B': 'Clean, Prepare, & Manipulate Data',
    'C': 'Train Model',
    'D': 'Test Data',
    'E': 'Improve'
}
G.add_nodes_from(node_labels.keys())

# 添加连接边
edges = [('A', 'B'), ('B', 'C'), ('C', 'D'), ('D', 'E')]
G.add_edges_from(edges)

# 设置布局（固定种子保证每次布局一致）
pos = nx.spring_layout(G, seed=42)

# 绘制节点、标签和边
nx.draw_networkx_nodes(G, pos, node_size=3000, node_color='lightblue')
nx.draw_networkx_labels(G, pos, labels=node_labels, font_size=10)
nx.draw_networkx_edges(G, pos, edgelist=edges, arrowstyle='->', arrowsize=20)

# 保存并显示图片
plt.axis('off')
plt.savefig('data_science_flowchart.png', dpi=300, bbox_inches='tight')
plt.show()

内容的提问来源于stack exchange，提问作者Vincent Granville