关于graphviz库的问题，报错png找不到，

incrediblejack

代码的功能主要是利用决策树进行分类，并画出决策树
我是用graphivz第三方库进行化的，但是报错了。报错如下：
pydotplus.graphviz.InvocationException: Program terminated with status: 1. stderr follows: Format: "png" not recognized. Use one of:

Process finished with exit code 1

也不给提示到底用什么格式的图片。
然而这个代码在我之前的旧电脑上是可以运行成功的，但是我换了个新电脑就报错了。我感觉应该是环境配置的问题。
我用的是anaconda的环境，在里面添加了graphviz库。
请各位大神帮帮忙咯！
谢谢

附上代码：

1. from sklearn.model_selection import train_test_split
   
2. from sklearn.tree import DecisionTreeClassifier, export_graphviz
   
3. import pydotplus
   
4. from IPython.display import display, Image
   
5. import pandas as pd
   
6. import numpy as np
   
7. 
   
8. fourier_feature_data = pd.read_csv('20200821_time_fre_01.csv',encoding='utf-8')  #读取数据集    #该打开方式需要先把原始的文件重新保存为utf-8形式的csv文件。#后面改成numpy.array试试
   
9. print('fourier_feature_data 数据类型 ： ', type(fourier_feature_data))
   
10. print(' ')
    
11. fourier_feature_data.columns = ['Apjs', 'Bjfg', 'Cfz', 'Dffz', 'Eqd','Ffzyz','Gmcyz','Hbxyz','Iydyz','Jpxd','Kf1','Lf2','Mf3','Nf4','Of5','Pf6','Qf7','Rf8','Sclass']  #表上数据集列名头
    
12. feature_name = fourier_feature_data.columns[:-1]
    
13. class_name = fourier_feature_data.columns[-1]
    
14. print(fourier_feature_data)
    
15. 
    
16. print(' ')
    
17. 
    
18. 
    
19. #下面是交叉验证
    
20. all_inputs = fourier_feature_data[['Apjs', 'Bjfg', 'Cfz', 'Dffz', 'Eqd','Ffzyz','Gmcyz','Hbxyz','Iydyz','Jpxd','Kf1','Lf2','Mf3','Nf4','Of5','Pf6','Qf7','Rf8']].values
    
21. all_classes = fourier_feature_data['Sclass'].values
    
22. 
    
23. (training_inputs, testing_inputs, training_classes, testing_classes) = train_test_split(all_inputs, all_classes, train_size = 0.70, random_state=1)
    
24. 
    
25. 
    
26. 
    
27. dtc = DecisionTreeClassifier(criterion='gini', max_depth=10)
    
28. dtc.fit(training_inputs,training_classes)
    
29. y_test_pre = dtc.predict(testing_inputs)
    
30. 
    
31. num = all_inputs.shape[0]
    
32. num_train = training_inputs.shape[0]
    
33. num_test = num - num_train
    
34. acc = sum(y_test_pre == testing_classes) / num_test
    
35. print('The accuracy is ', acc)
    
36. print(' ')
    
37. print('all inputs are : ', all_inputs)
    
38. print(' ')
    
39. print('all classes are : ', all_classes)
    
40. print(' ')
    
41. print('feature names are :', feature_name)
    
42. print(' ')
    
43. print('class names are : ', class_name)
    
44. print(' ')
    
45. 
    
46. print('feature_name 数据类型 ： ', type(feature_name))
    
47. print(' ')
    
48. print('class_name 数据类型 ： ', type(class_name))
    
49. print(' ')
    
50. 
    
51. 
    
52. print('all_inputs 数据类型 ： ', type(all_inputs))
    
53. print(' ')
    
54. print('all_classes 数据类型 ： ', type(all_classes))
    
55. class_name_2 = np.array(['0','1','2','3'])    #轴承实验台用
    
56. class_name_m = np.array(['0','1'])           #毛博士给的东方所的数据集用
    
57. # print('行数：', fourier_feature_data[:-1])
    
58. # print('列数：', fourier_feature_data[,:-1])
    
59. #dot_data = export_graphviz(dtc, out_file=None, feature_names = fourier_feature_data.columns[:-1], class_names = fourier_feature_data.columns[-1],filled = True, rounded =True, special_characters = True)
    
60. dot_data = export_graphviz(dtc, out_file=None, feature_names = feature_name, class_names = class_name_2, filled = True, rounded =True, special_characters = True)
    
61. #问题主要出在feature_names  和  class_names
    
62. 
    
63. graph = pydotplus.graph_from_dot_data(dot_data)
    
64. graph.write_png('20200821_jueceshu.png')
    
65. #abc = Image(graph.create_png())
    
66. image_path = 'C:/Users/rj/PycharmProjects/pythonProject/20200821_jueceshu.png'
    
67. Image(filename = image_path)
    
68. 
    
69. 
    

复制代码

附上数据集：
https://pan.baidu.com/s/1xpBb4Vtm2bqLPiZKKPN1EQ
提取码 a9xr

incrediblejack

求大神帮帮我咯！

incrediblejack

已解决，主要是环境配置的问题。具体环境配置随后我能贴图了的话，贴几个图就可以了。