print('coverged')
convergence = True
if i % 5000 == 0 or (convergence):
print('[{:4d}] loss : {:.4f} '.format(i, error))
if convergence:
break
# learning rate = 0.1
if __name__ == "__main__":
data_type = ['XOR', 'Linear']
for d in data_type:
X_train, y_train = GenData.fetch_data(d, 60)
X_test, y_test = GenData.fetch_data(d, 20)
net = SimpleNet([2, 4, 4, 1], learning_rate=0.1)
net.train(X_train, y_train, threshold=0.0001)
print('=' * 15, d, '=' * 15)
# Training (data leakage)
y_pred = net.forward(X_train)
print('train loss : ', loss(y_pred, y_train))
print('train accuracy : {:.2f}%'.format(
np.count_nonzero(np.round(y_pred) == y_train) * 100 / len(y_pred)))
print('-' * 30)
T_inf = 25
z = np.linspace(0, 1, N + 1)
# parameters for GenData
M = 32
# parameters for Prior
beta_prior = np.ones((N - 1, ))
sigma_prior = 1
C_beta = np.diag(sigma_prior**2 * np.ones((N - 1, )))
# start
model = Model(N, eps_0, h, T_inf)
prior = Prior(beta_prior, C_beta)
data = GenData(M, h, T_inf)
R = np.linalg.cholesky(C_beta)
N_samples = int(1e3)
beta_samp_lst = []
T_samp_lst = []
for i in range(N_samples):
s = np.random.randn(N - 1)
beta_sample = beta_prior + R.dot(s)
T_sample, _ = model.get_T_mult(beta_sample)
T_samp_lst.append(T_sample)
N = 40
h = 0.5
eps_0 = 5e-4
T_inf = 50
z = np.linspace(0, 1, N+1)
M = 20
beta_prior = np.ones((N-1,))
sigma_prior = 1e-1
C_beta = np.diag(sigma_prior**2*np.ones((N-1,)))
model = Model(N, eps_0, h, T_inf)
prior = Prior(beta_prior, C_beta)
data = GenData(M, h, T_inf)
data.gen_data(2, 'vector', scalar_noise=0.02)
objfn = ObjectiveFn(z, data, model, prior)
data_true = GenData(N, h, T_inf)
T_true, _ = data_true.get_T_true()
beta_test = model.get_beta_true(T_true)
G_adjoint = objfn.compute_gradient_adjoint(beta_test)
G_direct = objfn.compute_gradient_direct(beta_test)
G_adjoint_cont = objfn.compute_gradient_adjoint_cont(beta_test)
plt.figure()
for epsilon in [-1,-2,-3,-4]:
G_findiff = objfn.compute_gradient_findiff(beta_test, epsilon=10**epsilon)
N = 32 h = 0.5 eps_0 = 5e-4 T_inf = 50 * np.ones((N - 1, )) z = np.linspace(0, 1, N + 1) M = 32 T_inf_data = 50 * np.ones((M - 1, )) beta_prior = np.ones((N - 1, )) sigma_prior = 0.8 C_beta = np.diag(sigma_prior**2 * np.ones((N - 1, ))) model = Model(N, eps_0, h, T_inf) prior = Prior(beta_prior, C_beta) data = GenData(M, h, T_inf_data) data.gen_data(100, C_m_type=C_m_type, scalar_noise=0.02) objfn = ObjectiveFn(z, data, model, prior) beta0 = np.ones(N - 1) optimizer = 'CG' opt = Optimization(model, prior, data, objfn, beta0, optimizer) opt.compute_MAP_properties() opt.compute_base_properties() opt.sample_posterior(int(1e6)) opt.compute_true_properties() f = plt.figure(figsize=(14, 4))
import numpy as np
import matplotlib.pyplot as plt
from Models import Model
from GenData import GenData
N = 30
h = 0.5
eps_0 = 5e-4
z = np.linspace(0, 1, N + 1)
plt.figure()
for T_inf in np.linspace(5, 50, 5):
data = GenData(N, h, T_inf)
mod = Model(N, eps_0, h, T_inf)
T_true, _ = data.get_T_true()
T_base, _ = mod.get_T_base()
plt.plot(z[1:N],
T_true,
marker='s',
fillstyle='none',
linestyle='none',
color='k',
label='True' if T_inf == 5 else '')
plt.plot(z[1:N], T_base, 'r', label='Base' if T_inf == 5 else '')
plt.xlabel('z')