Question

我正使用Keras中的GridSearchCV调整sklearn模型中的超参数，如this tutorial

model = KerasClassifier(build_fn=create_model, verbose=0)
batch_size = [10, 20, 40, 60, 80, 100]
epochs = [10, 50, 100]
param_grid = dict(batch_size=batch_size, epochs=epochs)
grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1)

他们调用fit方法找到最佳超参数

grid_result = grid.fit(X, Y)

但是，让我们说我想要更改batch_sizes并再次调用fit（无需在Jupyter中重新启动内核）。

batch_size = [15, 20, 25, 30, 35, 40]
param_grid = dict(batch_size=batch_size, epochs=epochs)
grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1)

当我致电fit

时

grid_result = grid.fit(X, Y)

它无休止地工作并且不会终止。为了适应这些已更改的参数，我必须重新启动内核，然后重新加载数据，模块等。

问题。如何在不重新启动内核的情况下第二次在fit上调用GridSearchCV？

详细信息。我使用this data。详细摘录：

import numpy as np
from sklearn.model_selection import GridSearchCV
from keras.models import Sequential
from keras.layers import Dense
from keras.wrappers.scikit_learn import KerasClassifier

def create_model():
    model = Sequential()
    model.add(Dense(12, input_dim=8, activation='relu'))
    model.add(Dense(1, activation='sigmoid'))
    model.compile(loss='binary_crossentropy', optimizer='adam', metrics=['accuracy'])
    return model

dataset = np.loadtxt("data/pima-indians-diabetes.data.csv", delimiter=",")
X = dataset[:,0:8]
y = dataset[:,8]

model = KerasClassifier(build_fn=create_model, verbose=0)

batch_size = [10, 20, 40, 60, 80, 100]
epochs = [10, 50]
param_grid = dict(batch_size = batch_size, epochs = epochs)

grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1)

然后我调用fit并且它可以正常工作

grid_result = grid.fit(X, y)
print("Best: %f using %s" % (grid_result.best_score_, grid_result.best_params_))

出：Best: 0.690104 using {'batch_size': 10, 'epochs': 50}

然后我运行以下内容进行一些更改：

batch_size = [5, 10, 15, 20]
param_grid = dict(batch_size = batch_size, epochs = epochs)
grid = GridSearchCV(estimator=model, param_grid=param_grid, n_jobs=-1)

最后我第二次打电话给fit，永远不会停止