估计器MLPClassifier的参数估计器无效

Question

当我试图将GridSearchCV用于MLPClassifier时:

from sklearn.neural_network import MLPClassifier
from sklearn.model_selection import GridSearchCV
from sklearn.datasets import make_classification
from numpy import logspace
X, y = make_classification(random_state=42)
mlp = MLPClassifier(solver="lbfgs", hidden_layer_sizes=(100,), random_state=1)
paramgrid = {
    "estimator__alpha": logspace(-3, 2, 20),
}
mlpcv = GridSearchCV(mlp, paramgrid, cv=5)
mlpcv.fit(X, y)
print(mlpcv.best_params_)

但是我得到了这个消息:

ValueError: Invalid parameter 'estimator' for estimator 
  MLPClassifier(random_state=1, solver='lbfgs').
  Valid parameters are: ['activation', 'alpha', 'batch_size', 
  'beta_1', 'beta_2', 'early_stopping', 'epsilon', 
  'hidden_layer_sizes', 'learning_rate', 'learning_rate_init', 
  'max_fun', 'max_iter', 'momentum', 'n_iter_no_change', 
  'nesterovs_momentum', 'power_t', 'random_state', 'shuffle', 'solver', 
  'tol', 'validation_fraction', 'verbose', 'warm_start'].

哪个参数有错误?

Answer 1

您正在对包含参数estimator__alpha的paramgrid进行网格搜索。

但是MLPClassifier没有这个参数。你应该改变你的paramgrid .