class LogRegTests(unittest.TestCase):
def setUp(self):
self.stat_calc = Identity(degree=2, cross=False)
self.distancefunc = LogReg(self.stat_calc, seed=1)
self.rng = np.random.RandomState(1)
def test_distance(self):
d1 = 0.5 * self.rng.randn(100, 2) - 10
d2 = 0.5 * self.rng.randn(100, 2) + 10
d1 = d1.tolist()
d2 = d2.tolist()
# Checks whether wrong input type produces error message
self.assertRaises(TypeError, self.distancefunc.distance, 3.4, d2)
self.assertRaises(TypeError, self.distancefunc.distance, d1, 3.4)
# completely separable datasets should have a distance of 1.0
self.assertEqual(self.distancefunc.distance(d1, d2), 1.0)
# equal data sets should have a distance of 0.0
self.assertEqual(self.distancefunc.distance(d1, d1), 0.0)
def test_dist_max(self):
self.assertTrue(self.distancefunc.dist_max() == 1.0)
class LogRegTests(unittest.TestCase):
def setUp(self):
self.stat_calc = Identity(degree = 1, cross = 0)
self.distancefunc = LogReg(self.stat_calc)
def test_distance(self):
d1 = 0.5 * np.random.randn(100,2) - 10
d2 = 0.5 * np.random.randn(100,2) + 10
#Checks whether wrong input type produces error message
self.assertRaises(TypeError, self.distancefunc.distance, 3.4, d2)
self.assertRaises(TypeError, self.distancefunc.distance, d1, 3.4)
# completely separable datasets should have a distance of 1.0
self.assertEqual(self.distancefunc.distance(list(d1),list(d2)), 1.0)
# equal data sets should have a distance of 0.0
self.assertEqual(self.distancefunc.distance(list(d1),list(d1)), 0.0)
def test_dist_max(self):
self.assertTrue(self.distancefunc.dist_max() == 1.0)
distance_calculator = Euclidean(statistics_calculator)
distance_calculator2 = Wasserstein(statistics_calculator)
distance_calculator3 = LogReg(statistics_calculator)
# checking
sim_data = hh_simulator.forward_simulate(
[gbar_K_true + 3, gbar_Na_true + 4])
#obs_data = [obs_data[0] / np.max(obs_data[0]) + noise]
#sim_data = [sim_data[0] / np.max(sim_data[0])]
s1 = statistics_calculator.statistics(obs_data)
s2 = statistics_calculator.statistics(sim_data)
dist = distance_calculator.distance(obs_data, sim_data)
dist2 = distance_calculator2.distance(obs_data, sim_data)
dist3 = distance_calculator3.distance(obs_data, sim_data)
print(s1)
print(s2)
print(dist)
print(dist2)
print(dist3)
# plt.plot(t, obs_data[0] / np.max(obs_data[0]), label='obs')
# plt.plot(t, sim_data[0] / np.max(sim_data[0]), label='sim')
plt.plot(t, obs_data[0], label='obs')
plt.plot(t, sim_data[0], label='sim')
plt.legend()
plt.show()
'''
# inference scheme
sampler = RejectionABC(