def train_mlp():
with open('../data/params_0.pkl', 'rb') as f:
w_0, _, _ = cPickle.load(f)
with open('../data/params_1.pkl', 'rb') as f:
w_1, _, _ = cPickle.load(f)
with open('../data/params_2.pkl', 'rb') as f:
w_2, _, _ = cPickle.load(f)
train_x, train_y = SupervisedLoader.load('../data')
model = Sequential()
model.add(Dense(33, 64, weights=[w_0]))
model.add(Activation('sigmoid'))
# model.add(Dropout(0.2))
model.add(Dense(64, 128, weights=[w_1]))
model.add(Activation('sigmoid'))
# model.add(Dropout(0.2))
model.add(Dense(128, 128, weights=[w_2]))
model.add(Dense(128, 1, init='glorot_uniform'))
model.add(Activation('relu'))
# sgd = SGD(lr=1.e-5, decay=1e-6, momentum=0.9, nesterov=True)
model.compile(loss='mean_squared_error', optimizer='adagrad')
model.fit(train_x, train_y, nb_epoch=500, batch_size=128, validation_split=0.2)
model.save_weights('../data/mlp_params.hdf5')
def compute_features_from_aes_for_train_set():
train_x, train_y = SupervisedLoader.load('../data')
config = load_configuration('../config/caes.json')
scaes = StackedAutoencoders(config, warm_start=True)
train_x = scaes.get_features(train_x)
np.save('../data/features.npy', train_x)
np.save('../data/hazards.npy', train_y)
def grid_search_for_svr():
train_x, train_y = SupervisedLoader.load('../data')
gammas = [4.]
clf = SVR(verbose=1)
param_grid = {'gamma': gammas, 'C': [10., 20., 30., 40.]}
grid_search = GridSearchCV(clf, param_grid, scoring='mean_squared_error', n_jobs=4, verbose=1)
grid_search.fit(train_x, train_y)
print grid_search.best_score_
print grid_search.best_params_
with open('../data/another2_svr.pkl', 'wb') as f:
cPickle.dump(grid_search.best_estimator_, f)
