def cnn():
from keras.preprocessing.sequence import pad_sequences
from keras.wrappers.scikit_learn import KerasRegressor
from keras.models import Sequential
from keras.layers import Dropout, Dense, Activation, Conv2D, MaxPooling2D, Flatten, Conv1D, MaxPooling1D, \
GlobalAveragePooling1D, BatchNormalization
x, y = load_preprocessed_data_23_7()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=seed)
scaler = StandardScaler()
scaler.fit(x)
x_train = scaler.transform(x_train)
x_test = scaler.transform(x_test)
x_train = x_train.reshape(x_train.shape[0], 7, 7, 1)
x_test = x_test.reshape(x_test.shape[0], 7, 7, 1)
model = Sequential()
# model.add(BatchNormalization(input_shape=(7, 7, 1)))
model.add(
Conv2D(32, (3, 3),
input_shape=(7, 7, 1),
activation='relu',
padding='same'))
model.add(Conv2D(32, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Conv2D(64, (3, 3), activation='relu', padding='same'))
model.add(Conv2D(64, (3, 3), activation='relu', padding='same'))
model.add(MaxPooling2D(pool_size=(2, 2)))
model.add(Dropout(0.25))
model.add(Flatten())
model.add(Dense(256, activation='relu'))
model.add(Dropout(0.5))
model.add(Dense(1, activation='sigmoid'))
model.compile(optimizer='adam',
loss='binary_crossentropy',
metrics=['accuracy'])
print('begin')
model.fit(x_train,
y_train,
batch_size=30,
epochs=10,
validation_split=0.2,
shuffle=True)
result = model.predict(x_test)
metrics(result, y_test)
'''
def knn():
x, y = load_preprocessed_data_23_7()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=seed)
model = Pipeline([('Scaler', StandardScaler()),
('KNN', KNeighborsRegressor())])
model.fit(x_train, y_train)
result = model.predict(x_test)
metrics(result, y_test)
'''
def svr():
x, y = load_preprocessed_data_23_7()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=seed)
model = Pipeline([('Scaler', StandardScaler()),
('SVM', SVR(verbose=True))])
model.fit(x_train, y_train)
result = model.predict(x_test)
metrics(result, y_test)
'''
def gbr():
x, y = load_preprocessed_data_23_7()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=seed)
model = Pipeline([('Scaler', StandardScaler()),
('pca', PCA(random_state=seed)),
('RBR', GradientBoostingRegressor(random_state=seed))])
model.fit(x_train, y_train)
result = model.predict(x_test)
metrics(result, y_test)
'''
def eval_ensemble_model():
x, y = load_preprocessed_data_23_7()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=seed)
ensembles = dict()
ensembles['ScaledAB'] = Pipeline([('Scaler', StandardScaler()),
('AB',
AdaBoostRegressor(random_state=seed))])
# ensembles['ScaledAB-KNN'] = Pipeline([('Scaler',
# StandardScaler()),
# ('ABKNN',
# AdaBoostRegressor(KNeighborsRegressor(),
# random_state=seed))])
ensembles['ScaledAB-LR'] = Pipeline([
('Scaler', StandardScaler()),
('ABLR', AdaBoostRegressor(LinearRegression(), random_state=seed))
])
ensembles['ScaledRFR'] = Pipeline([
('Scaler', StandardScaler()),
('RFR', RandomForestRegressor(random_state=seed))
])
ensembles['ScaledETR'] = Pipeline([
('Scaler', StandardScaler()),
('ETR', ExtraTreesRegressor(random_state=seed))
])
ensembles['ScaledGBR'] = Pipeline([
('Scaler', StandardScaler()),
('RBR', GradientBoostingRegressor(random_state=seed))
])
results = []
for key in ensembles:
kfold = KFold(n_splits=num_folds, random_state=seed)
cv_result = cross_val_score(ensembles[key],
x_train,
y_train,
cv=kfold,
scoring=scoring,
n_jobs=4)
results.append(cv_result)
print('%s: %f (%f)' % (key, cv_result.mean(), cv_result.std()))
def eval_model():
x, y = load_preprocessed_data_23_7()
x_train, x_test, y_train, y_test = train_test_split(x,
y,
test_size=0.2,
random_state=seed)
pipelines = dict()
pipelines['ScalerLR'] = Pipeline([('Scaler', StandardScaler()),
('LR',
LogisticRegression(random_state=seed))])
pipelines['ScalerLinearR'] = Pipeline([('Scaler', StandardScaler()),
('LR', LinearRegression())])
pipelines['ScalerLASSO'] = Pipeline([('Scaler', StandardScaler()),
('LASSO', Lasso(random_state=seed))])
pipelines['ScalerEN'] = Pipeline([('Scaler', StandardScaler()),
('EN', ElasticNet(random_state=seed))])
# pipelines['ScalerKNN'] = Pipeline(
# [('Scaler', StandardScaler()), ('KNN', KNeighborsRegressor())])
pipelines['ScalerCART'] = Pipeline([
('Scaler', StandardScaler()),
('CART', DecisionTreeRegressor(random_state=seed))
])
pipelines['ScalerSVM'] = Pipeline([('Scaler', StandardScaler()),
('SVM', SVR())])
results = []
for key in pipelines:
kfold = KFold(n_splits=num_folds, random_state=seed)
cv_result = cross_val_score(pipelines[key],
x_train,
y_train,
cv=kfold,
scoring=scoring,
n_jobs=4)
results.append(cv_result)
print('%s: %f (%f)' % (key, cv_result.mean(), cv_result.std()))