GridSearchCV, deep learning

!pip install -U keras-tuner
import kerastuner as kt

import numpy as np
from tensorflow.keras.models import Sequential
from tensorflow.keras.layers import Dense, Dropout, Flatten
import keras
import tensorflow as tf
import IPython
from tensorflow.keras.wrappers.scikit_learn import KerasClassifier
from sklearn.model_selection import GridSearchCV

# 모델 만들기
tf.random.set_seed(7)

def model_builder(nodes=16, activation='relu'):

  model = Sequential()
  model.add(Dense(nodes, activation=activation))
  model.add(Dense(nodes, activation=activation))
  model.add(Dense(1, activation='sigmoid')) # 이진분류니까 노드수 1, 활성함수로는 시그모이드

  model.compile(optimizer='adam', 
                loss='binary_crossentropy', 
                metrics=['accuracy'])

  return model
  
  
#keras.wrapper 분류기
model = KerasClassifier(build_fn=model_builder, verbose=0)

# GridSearch 파라미터 설정
batch_size = [50, 100, 300]
epochs = [10, 20, 30]
nodes = [64, 128, 256]
activation = ['relu', 'sigmoid']
param_grid = dict(batch_size=batch_size, epochs=epochs, nodes=nodes, activation=activation)


# GridSearch CV를 만들기
grid = GridSearchCV(estimator=model, param_grid=param_grid, cv=3, verbose=1, n_jobs=-1)
grid_result = grid.fit(X_train_scaled, y_train)

# 최적의 결과값을 낸 파라미터를 출력합니다
print(f"Best: {grid_result.best_score_} using {grid_result.best_params_}")
means = grid_result.cv_results_['mean_test_score']
stds = grid_result.cv_results_['std_test_score']
params = grid_result.cv_results_['params']
for mean, stdev, param in zip(means, stds, params):
    print(f"Means: {mean}, Stdev: {stdev} with: {param}")

##그리드 서치에서 결괏값을 알려주는 코드