ValueError：至少需要一个数组才能与sklearn cross_val_predict方法连接

如何解决ValueError：至少需要一个数组才能与sklearn cross_val_predict方法连接

我正在尝试使用SVM分类器通过自定义交叉验证折叠对二进制分类问题进行建模，但是它给我带来了错误**需要至少一个数组以将cross_val_predict连接在一起**。该代码在cros_val_predict中的cv = 3上可以正常工作，但是当我使用custom_cv时，会出现此错误。

下面是代码：

from sklearn.model_selection import LeavePOut
import numpy as np
from sklearn.svm import SVC
from time import *
from sklearn.metrics import roc_auc_score
from sklearn.model_selection import cross_val_predict,cross_val_score
clf = SVC(kernel='linear',C=25)
X = np.array([[1,2],[3,4],[5,6],[7,8],[9,10]])
y = np.array([0,1,0])
lpo = LeavePOut(2)
print(lpo.get_n_splits(X))
LeavePOut(p=2)
test_index_list=[]
train_index_list=[]
for train_index,test_index in lpo.split(X,y):
  
  if(y[test_index[0]]==y[test_index[1]]):
    pass
  else:
    print("TRAIN:",train_index,"TEST:",test_index)
    X_train,X_test = X[train_index],X[test_index]
    y_train,y_test = y[train_index],y[test_index]
    train_index_list.append(train_index)
    test_index_list.append(test_index)
custom_cv = zip(train_index_list,test_index_list)
scores = cross_val_score(clf,X,y,cv=custom_cv)

print(scores)
print('accuracy:',scores.mean())
predicted=cross_val_predict(clf,cv=custom_cv) # error with this line
print('Confusion matrix:',confusion_matrix(labels,predicted))

下面是完整的错误痕迹：

ValueError                                Traceback (most recent call last)
<ipython-input-11-d78feac932b2> in <module>()
     31 print(scores)
     32 print('accuracy:',scores.mean())
---> 33 predicted=cross_val_predict(clf,cv=custom_cv)
     34 
     35 print('Confusion matrix:',predicted))

/usr/local/lib/python3.6/dist-packages/sklearn/model_selection/_validation.py in cross_val_predict(estimator,groups,cv,n_jobs,verbose,fit_params,pre_dispatch,method)
    758     predictions = [pred_block_i for pred_block_i,_ in prediction_blocks]
    759     test_indices = np.concatenate([indices_i
--> 760                                    for _,indices_i in prediction_blocks])
    761 
    762     if not _check_is_permutation(test_indices,_num_samples(X)):

<__array_function__ internals> in concatenate(*args,**kwargs)

ValueError: need at least one array to concatenate

关于如何解决此错误的任何建议？

解决方法

这里有2个错误：

1. 如果要重用zip对象，请从中创建一个列表。使用一次后，该对象将耗尽。您可以像这样修复它：

custom_cv = [*zip(train_index_list,test_index_list)]

cross_val_predict的
1. 交叉验证列表应该是实际数组（Each sample should only belong to exactly one test set）的分区。在您的情况下不是。如果考虑一下，从交叉验证列表中堆叠输出将得到长度为 6 的数组，而原始的 y 则为长度5。您可以像这样实现自定义交叉值预测：

def custom_cross_val_predict(clf,X,y,cv):
    y_pred,y_true = [],[]
    for tr_idx,vl_idx in cv:
        X_tr,y_tr = X[tr_idx],y[tr_idx]
        X_vl,y_vl = X[vl_idx],y[vl_idx]
        clf.fit(X_tr,y_tr)
        y_true.extend(y_vl)
        y_pred.extend(clf.predict(X_vl))
        
    return y_true,y_pred

labels,predicted = custom_cross_val_predict(clf,cv=custom_cv)
print('Confusion matrix:',confusion_matrix(labels,predicted))