Python绘制流程图，插图图[关闭]

Question

有没有python库可以画流程图，图解图如：

Answer 1

您可以使用 Schemadraw。虽然它的主要目的是制作高质量的电路原理图，但也有一部分可以用来绘制流程图。

例如： https://schemdraw.readthedocs.io/en/latest/gallery/flowcharting.html#galleryflow

Answer 2

Network Charts 可能会成功。查看Networkx docs 了解更多详细信息。这也是为大型网络设计的，但是如果您结合其中的一些示例，可以对其进行一些定制以用作流程图。我可以通过一点挖掘来创建它，它可以作为流程图的一个不错的模板。

import pandas as pd
import networkx as nx
import matplotlib.pyplot as plt

plt.figure(figsize = (12,9))

From = ['Food\nProduction', 'Transportation', 'Energy\nProduction',
        "Greenhouse\nGas\nEmissions",'Climate\nChange','Climate\nFeedbacks','Greenhouse\nGas\nEmissions',
        'Climate\nChange']
To = ["Greenhouse\nGas\nEmissions", "Greenhouse\nGas\nEmissions",
      "Greenhouse\nGas\nEmissions",'Climate\nChange','Climate\nFeedbacks','Greenhouse\nGas\nEmissions',
      'Climate\nChange','Everyone$^{Dies}$']

df = pd.DataFrame({ 'from':From,
                   'to':To})
# Define Node Positions
pos = {'Food\nProduction':(1,1),
        'Transportation':(1,2),
        'Energy\nProduction':(1,3),
        'Greenhouse\nGas\nEmissions':(2,2),
        'Climate\nChange':(3,2),
        'Climate\nFeedbacks':(2.5,3),
        'Everyone$^{Dies}$':(4,2)}

# Define Node Colors
NodeColors = {'Food\nProduction':[1,0,1],
        'Transportation':[1,1,0],
        'Energy\nProduction':[0,1,1],
        'Greenhouse\nGas\nEmissions':[1,.5,1],
        'Climate\nChange':[0,1,0],
        'Climate\nFeedbacks':[0,0,1],
        'Everyone$^{Dies}$':[1,0,0]}

Labels = {}
i = 0
for a in From:
    Labels[a]=a
    i +=1
Labels[To[-1]]=To[-1]

# Build your graph. Note that we use the DiGraph function to create the graph! This adds arrows
G=nx.from_pandas_edgelist(df, 'from', 'to', create_using=nx.DiGraph() )

# Define the colormap and set nodes to circles, but the last one to a triangle
Circles = []
Traingle = []
Colors_Circles = []
Colors_Traingle = []
for n in G.nodes:
    if n != 'Everyone$^{Dies}$':
        Circles.append(n)
        Colors_Circles.append(NodeColors[n])
    else:
        Traingle.append(n)
        Colors_Traingle.append(NodeColors[n])

# By making a white node that is larger, I can make the arrow "start" beyond the node
nodes = nx.draw_networkx_nodes(G, pos, 
                       nodelist = Circles,
                       node_size=1.25e4,
                       node_shape='o',
                       node_color='white',
                       alpha=1)

nodes = nx.draw_networkx_nodes(G, pos, 
                       nodelist = Circles,
                       node_size=1e4,
                       node_shape='o',
                       node_color=Colors_Circles,
                       edgecolors='black',
                       alpha=0.5)


nodes = nx.draw_networkx_nodes(G, pos, 
                       nodelist = Traingle,
                       node_size=1.25e4,
                       node_shape='>',
                       node_color='white',
                       alpha=1)

nodes = nx.draw_networkx_nodes(G, pos, 
                       nodelist = Traingle,
                       node_size=1e4,
                       node_shape='>',
                       node_color=Colors_Traingle,
                       edgecolors='black',
                       alpha=0.5)


nx.draw_networkx_labels(G, pos, Labels, font_size=12)

# Again by making the node_size larer, I can have the arrows end before they actually hit the node
edges = nx.draw_networkx_edges(G, pos, node_size=1.8e4,
                               arrowstyle='->',width=2,arrowsizes=10)

plt.xlim(0,4.5)
plt.ylim(0,4)
plt.axis('off')
plt.show()

Answer 3

您可能想试试 PyDot： https://code.google.com/p/pydot/

python版本没试过，不过之前用过Graphviz，挺好用的。