使用一个热编码器时出现gridsearchCV错误

Question

我在使用带有一种热编码的 gridsearch cv 时遇到此错误： “分类指标无法处理多标签指标和多类目标的混合”

我的 y_train 形状是：(64345, 37)，我的 X_train 形状是：(64345, 14)。

我无法弄清楚我哪里出错了。任何指导/帮助将不胜感激。

它可以为我的模型正确执行，而无需使用带有固定参数的 gridsearchCV。 如果不使用一种热编码，我会得到索引超出范围的错误。 该帖子的链接在这里：I was training an Ann machine learning model using GridSearchCV and got stuck with an IndexError in gridSearchCV

这是我拆分数据集的方法：

from sklearn.preprocessing import LabelEncoder, OneHotEncoder

onehotencoder = OneHotEncoder(categorical_features = [0])
df = onehotencoder.fit_transform(df).toarray()
df=df[:,1:]

target=df[:,0:37]
dataset=df[:,37:51]
from sklearn.model_selection import train_test_split
X_train,X_test,y_train,y_test=train_test_split(dataset,target,random_state=1)
from sklearn.preprocessing import StandardScaler
sc = StandardScaler()
X_train= sc.fit_transform(X_train)
X_test=pd.DataFrame(X_test) 

这里是 gridseachcv 代码：

from keras.wrappers.scikit_learn import KerasClassifier
from sklearn.model_selection import GridSearchCV
from keras.models import Sequential
from keras.layers import Dense
def build_classifier(optimizer, nb_layers,unit):
    classifier = Sequential()
    classifier.add(Dense(units = unit, kernel_initializer = 'uniform', activation = 'relu', input_dim = 14))
    i = 1
    while i <= nb_layers:
        classifier.add(Dense(activation="relu", units=unit, kernel_initializer="uniform"))
        i += 1
    classifier.add(Dense(units = 37, kernel_initializer = 'uniform', activation = 'softmax'))
    classifier.compile(optimizer = optimizer, loss = 'categorical_crossentropy', metrics = ['accuracy'])
    return classifier
classifier = KerasClassifier(build_fn = build_classifier)
parameters = {'batch_size': [10,25,32,64,128,256],
              'epochs': [50,100, 200,500,1000,1500,2000],
              'optimizer': ['adam'],
              'nb_layers': [2,3,4,5,6],
              'unit':[28,40,48,57]
             }
grid_search = GridSearchCV(estimator = classifier,
                           param_grid = parameters,
                           scoring = 'accuracy',
                          cv=10,n_jobs=-1)
grid_search = grid_search.fit(X_train, y_train)
best_parameters = grid_search.best_params_
best_accuracy = grid_search.best_score_

我应该得到最好的结果参数，但我得到了错误- ValueError：分类指标无法处理多标签指标和多类目标的混合

Answer 1

错误信息很清楚。

在这里，您有y_train:(64345, 37)，这意味着每个样本都是多标签的。每个样本有 37 个标签。

sklearn的分类指标无法处理多标签的目标变量。

在申请GridSearch()之前，您应该寻找获得y_train:(64345, 1)的方法

---

对于可以处理红色多标签问题的模型：

https://scikit-learn.org/stable/modules/multiclass.html

Support multilabel:

sklearn.tree.DecisionTreeClassifier
sklearn.tree.ExtraTreeClassifier
sklearn.ensemble.ExtraTreesClassifier
sklearn.neighbors.KNeighborsClassifier
sklearn.neural_network.MLPClassifier
sklearn.neighbors.RadiusNeighborsClassifier
sklearn.ensemble.RandomForestClassifier
sklearn.linear_model.RidgeClassifierCV