为什么在使用 LightGBM 和 Sklearn 完成多分类任务时出现“Python int too large to convert to C long”？

如何解决为什么在使用 LightGBM 和 Sklearn 完成多分类任务时出现“Python int too large to convert to C long”？

我想用LightGBM和Sklearn完成一个多分类任务，代码如下

#Input
X = copy.deepcopy(GTEx_feature)
y = copy.deepcopy(out)
#TrainingSet : TestSet = 4 : 1
X_train,X_test,y_train,y_test = train_test_split(X,y,test_size = 0.2,random_state = SEED)
########################################
########## Model Construction ##########
print("\n...... Training Model ......\n")
# LightGBM params
param_dict = {
    "objective":'multiclass',"num_class":[5],"learning_rate":[0.1,0.05,0.02,0.015,0.01],#
    "num_leaves": range(10,36,5),# Maximum tree leaves for base learners.
    "max_depth" : [-1,2,3,4,5,10,20,40,50],#Maximum tree depth for base learners,<=0 means no limit.
    "min_data_in_leaf": range(1,45,2),#
    "feature_fraction" : [i / 10 for i in range(2,11)],#
    "metric" : "multi_error",#
    "early_stopping_rounds" : [None],#
    "n_jobs" : [-1],#
    "silent" : [True],#
    "verbose" : [-1],#
    "n_estimators" : range(50,1000,50),#
    "bagging_fraction" :  [i / 10 for i in range(2,#
    "bagging_freq" : [0,1,2],#
    "lambda_l1" : [0,0.001,0.005,0.01,0.1],#
    "lambda_l2" : [0,#
    "random_state":[2020]#
}
#Initiate model
model = lgb.LGBMClassifier()
#Adjust hyper-parameters with 5-fold cross validation
rscv = RandomizedSearchCV(model,#
                          param_dict,#
                          n_iter=100,# Number of parameter settings that are sampled. n_iter Tradesoff runtime vs quality of the solution.
                          cv = 5,# Determines the cross-validation splitting strategy.
                          verbose = 0,# Controls the verbosity: the higher,the more messages.
                          scoring = "roc_auc",#
                          n_jobs =-1# 
                         )
gbm=rscv.fit(X_train,y_train)

最后一行出现错误：

---------------------------------------------------------------------------
OverflowError                             Traceback (most recent call last)
<ipython-input-110-97f7932a1488> in <module>
     42                           n_jobs =-1#
     43                          )
---> 44 gbm=rscv.fit(X_train,y_train)
     45 ########## Model Evaluation ##########
     46 print("\n...... Evaluating Model ......\n")

~\anaconda3\lib\site-packages\sklearn\utils\validation.py in inner_f(*args,**kwargs)
     71                           FutureWarning)
     72         kwargs.update({k: arg for k,arg in zip(sig.parameters,args)})
---> 73         return f(**kwargs)
     74     return inner_f
     75 

~\anaconda3\lib\site-packages\sklearn\model_selection\_search.py in fit(self,X,groups,**fit_params)
    734                 return results
    735 
--> 736             self._run_search(evaluate_candidates)
    737 
    738         # For multi-metric evaluation,store the best_index_,best_params_ and

~\anaconda3\lib\site-packages\sklearn\model_selection\_search.py in _run_search(self,evaluate_candidates)
   1527     def _run_search(self,evaluate_candidates):
   1528         """Search n_iter candidates from param_distributions"""
-> 1529         evaluate_candidates(ParameterSampler(
   1530             self.param_distributions,self.n_iter,1531             random_state=self.random_state))

~\anaconda3\lib\site-packages\sklearn\model_selection\_search.py in evaluate_candidates(candidate_params)
    698 
    699             def evaluate_candidates(candidate_params):
--> 700                 candidate_params = list(candidate_params)
    701                 n_candidates = len(candidate_params)
    702 

~\anaconda3\lib\site-packages\sklearn\model_selection\_search.py in __iter__(self)
    282                     % (grid_size,grid_size),UserWarning)
    283                 n_iter = grid_size
--> 284             for i in sample_without_replacement(grid_size,n_iter,285                                                 random_state=rng):
    286                 yield param_grid[i]

sklearn\utils\_random.pyx in sklearn.utils._random.sample_without_replacement()

OverflowError: Python int too large to convert to C long

所以我想知道为什么会发生这个错误以及我可以尝试什么来处理它

这是一行X

g                              1.057671 
c                            2.644094  
fi                            -0.302407  
ph                            -0.772771  
da                            -0.449314  
phy                           -0.447774  
gen                           -1.042650  
gs                             0.053665  
ts                            -0.197370  
sig                           -0.137325  
endna                         -0.255032  
eireig                        -0.372373  
ee3                           -0.159200  
cong                          0.000000  
fhft                          0.000000  
sple                          0.000000  
p                             0.000000  
ts                            0.000000  
kb                            0.000000  
non                           1.000000  
sn                            1.000000  
nss                          -0.014290 
Name: 1,dtype: float64

在使用相同的 X 完成二进制分类时，我没有遇到这样的错误， y如下

0       2
1       0
2       0
3       4
4       4
       ..
4995    0
4996    1
4997    1
4998    0
4999    0
Name: mtle,Length: 5000,dtype: int64