我的模型准确率为什么那么低，是哪有问题吗，随机森林不是准确率很高吗

嘴角向上

输出结果0.37010975031949667 9  不应该啊

1. import pandas as pd #导入数据集
   
2. url=r"C:\Users\zmj佳佳佳\Desktop\第六步离散化测试.csv"
   
3. df = pd.read_csv(url, header = None,low_memory=False)#将数据集分为训练集和测试集
   
4. df.columns=["grade","dti","delinq_2yrs","earliest_cr_line","fico_range_low","inq_last_6mths",
   
5.             "mths_since_last_delinq","pub_rec","revol_bal","revol_util","mths_since_last_major_derog",
   
6.             "tot_cur_bal","open_acc_6m","open_il_12m","open_il_24m","mths_since_rcnt_il","open_rv_12m",
   
7.             "open_rv_24m","max_bal_bc","all_util","inq_last_12m","acc_open_past_24mths","avg_cur_bal",
   
8.             "bc_open_to_buy","mo_sin_old_il_acct","mo_sin_old_rev_tl_op","mo_sin_rcnt_rev_tl_op","mo_sin_rcnt_tl",
   
9.            "mort_acc","mths_since_recent_bc_dlq","mths_since_recent_inq","mths_since_recent_revol_delinq",
   
10.             "num_accts_ever_120_pd","num_actv_bc_tl","num_actv_rev_tl","num_bc_sats","num_bc_tl",
    
11.             "num_rev_accts","num_rev_tl_bal_gt_0","num_tl_90g_dpd_24m","num_tl_op_past_12m","pct_tl_nvr_dlq",
    
12.             "pub_rec_bankruptcies"]
    
13. #将数据集分为训练集和测试集
    
14. from sklearn.model_selection import train_test_split
    
15. from sklearn.model_selection import GridSearchCV
    
16. from sklearn.ensemble import RandomForestClassifier
    
17. x, y = df.iloc[:, 1:].values, df.iloc[:, 0].values
    
18. x_train, x_test, y_train, y_test = train_test_split(x, y, test_size = 0.3, random_state = 0)
    
19. feat_labels = df.columns[1:]
    
20. #param={"n_estimators":[10,20],"max_depth":[5,8]}
    
21. #网格搜索与交叉验证
    
22. #gc=GridSearchCV(forest,param_grid=param,cv=3)
    
23. #gc.fit(x_train,y_train)
    
24. #print("准确率：",gc.score(x_test,y_test))
    
25. #print("查看选择的参数模型：",gc.best_params_)
    
26. 
    
27. 
    
28. #n_estimators的学习曲线
    
29. 
    
30. from sklearn.model_selection import cross_val_score
    
31. import matplotlib.pyplot as plt
    
32. superpa = []
    
33. for i in range(0,10):
    
34.     forest = RandomForestClassifier(n_estimators=i+1,n_jobs=-1)
    
35.     rfc_s = cross_val_score(forest,x,y,cv=2).mean()
    
36.     superpa.append(rfc_s)
    
37. print(max(superpa),superpa.index(max(superpa)))
    
38. plt.figure(figsize=[20,5])
    
39. plt.plot(range(1,11),superpa)
    
40. plt.show()
    
41. 
    
42. 
    
43. 
    

复制代码

嘴角向上

有没有人可以帮忙指导下吗