想问下处理数据集不平衡问题的代码怎么写

嘴角向上

应该是我的数据集不平衡，用代码要怎么处理，处理之后我想用随机森林建模，所以想知道在模型
前面怎么加代码来处理数据集的不平衡问题

1. import pandas as pd  
   
2. from sklearn.model_selection import train_test_split
   
3. from sklearn.ensemble import RandomForestClassifier
   
4. import numpy as np
   
5. from sklearn.metrics import accuracy_score
   
6. from sklearn.metrics import confusion_matrix
   
7. url=r"C:\Users\zmj佳佳佳\Desktop\测试.csv"
   
8. df = pd.read_csv(url, header=None, low_memory=False)  
   
9. df.columns = ["grade", "dti", "delinq_2yrs", "earliest_cr_line", "fico_range_low", "inq_last_6mths",
   
10.               "mths_since_last_delinq", "pub_rec", "revol_bal", "revol_util", "mths_since_last_major_derog",
    
11.               "tot_cur_bal", "open_acc_6m", "open_il_12m", "open_il_24m", "mths_since_rcnt_il", "open_rv_12m",
    
12.               "open_rv_24m", "max_bal_bc", "all_util", "inq_last_12m", "acc_open_past_24mths", "avg_cur_bal",
    
13.               "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",
    
14.               "mort_acc", "mths_since_recent_bc_dlq", "mths_since_recent_inq", "mths_since_recent_revol_delinq",
    
15.               "num_accts_ever_120_pd", "num_actv_bc_tl", "num_actv_rev_tl", "num_bc_sats", "num_bc_tl",
    
16.               "num_rev_accts", "num_rev_tl_bal_gt_0", "num_tl_90g_dpd_24m", "num_tl_op_past_12m", "pct_tl_nvr_dlq",
    
17.               "pub_rec_bankruptcies"]
    
18. df.info()
    
19. # 将数据集分为训练集和测试集
    
20. x, y = df.iloc[:, 1:].values, df.iloc[:, 0].values
    
21. x_train, x_test, y_train, y_test = train_test_split(x, y, test_size=0.3, random_state=0)
    
22. feat_labels = df.columns[1:]
    
23. forest = RandomForestClassifier(random_state=0,n_jobs=-1, class_weight="balanced")
    
24. forest.fit(x_train, y_train)
    
25. 
    
26. print("准确率：", forest.score(x_train, y_train))
    
27. predict_results=forest.predict(x_test)
    
28. print(accuracy_score(predict_results, y_test))
    
29. 
    
30. print(confusion_matrix(predict_results, y_test))
    
31. 
    

复制代码

嘴角向上

因为我得模型准确率特别低，应该是数据集不平衡，我的目标变量如下：分为7类，这应该是不平衡才导致我得随机森林模型准确率只有40%吧                               
                次数       
有效        1        35494       
        2        30911       
        3        24827       
        4        13356       
        5        2962       
        6        23       
        7        14       
        总計        107587