kernel_regularizer=0.001)
model.add(1, kernel_initializer=1 / np.sqrt(4), kernel_regularizer=0.001)
# model.set_optimizer(
# SGD(
# lr = 0.8,
# momentum = 0.9,
# nesterov = True
# ))
# Batch
# model.set_optimizer(
# NCG()
# )
model.set_optimizer(LBFGS(m=20, c1=1e-4, c2=0.9, tol=1e-20))
model.fit(
X_train,
Y_train,
epochs=50,
#batch_size=31,
validation_data=[X_test, Y_test],
verbose=1)
outputNet = model.predict(X_test)
printMSE(outputNet, Y_test, type="test")
printAcc(outputNet, Y_test, type="test")
plotHistory(model.history)
print("Build the model")
model = Mlp()
model.add(4, input=17, kernel_initializer=0.003, kernel_regularizer=reg)
model.add(1, kernel_initializer=0.003, kernel_regularizer=reg)
#############################
# L-BFGS
#############################
c1 = 1e-4
c2 = .9
m = 30
ln_maxiter = 100
#############################
optimizer = LBFGS(m=m,
c1=c1,
c2=c2,
ln_maxiter=ln_maxiter,
norm_g_eps=ng_eps,
l_eps=l_eps)
model.set_optimizer(optimizer)
print("Start the optimization process:")
model.fit(X_train, Y_train, epochs=max_iter, verbose=verbose)
f = model.history["loss_mse_reg"]
##############################
# plot
##############################
pos_train = (0, 0)
figsize = (12, 4)
plt.plot(f - f[-1], linestyle='-')
model.add(4, input= 17, kernel_initializer = 0.003, kernel_regularizer = 0.001)
model.add(1, kernel_initializer = 0.003, kernel_regularizer = 0.001)
# model.set_optimizer(
# SGD(
# lr = 0.8,
# momentum = 0.6,
# nesterov = True
# ))
# model.set_optimizer(
# NCG(tol=1e-20)
# )
model.set_optimizer(
LBFGS(m=3, c1= 1e-4, c2=0.4, tol=1e-20)
)
# Batch
model.fit(X_train,
Y_train,
epochs=30,
#batch_size=31,
validation_data = [X_test, Y_test],
verbose=1)
outputNet = model.predict(X_test)
printMSE(outputNet, Y_test, type = "test")
printAcc(outputNet, Y_test, type = "test")
plotHistory(model.history )