def test_fit_sirc(self):
model = CovidSIRDc()
X, y = model.iterate2array(derivatives=True)
epi = EpidemicRegressor('sirc', max_iter=10)
with self.assertRaises(RuntimeError):
epi.score(X, y)
with self.assertRaises(RuntimeError):
epi.predict(X)
epi.fit(X, y)
loss = epi.score(X, y)
self.assertGreater(loss, 0)
def find_best_model(Xt, yt, lrs, th):
best_est, best_loss = None, None
for lr in lrs:
with warnings.catch_warnings():
warnings.simplefilter("ignore", RuntimeWarning)
m = EpidemicRegressor(
'SIR', learning_rate_init=lr,
max_iter=500, early_th=1)
m.fit(Xt, yt)
loss = m.score(Xt, yt)
if numpy.isnan(loss):
continue
if best_est is None or best_loss > loss:
best_est = m
best_loss = loss
if best_loss < th:
return best_est, best_loss
return best_est, best_loss
def test_clone(self):
model = CovidSIRD()
model['beta'] = 0.4
X, y = model.iterate2array(derivatives=True)
epi = EpidemicRegressor(max_iter=10)
epi.fit(X, y)
self.assertGreater(epi.iter_, 0)
pred1 = epi.predict(X)
f = io.BytesIO()
pickle.dump(epi, f)
epi2 = pickle.load(io.BytesIO(f.getvalue()))
pred2 = epi2.predict(X)
assert_almost_equal(pred1, pred2)
ds, cs = epi2.predict_many(X)
self.assertEqual(ds.shape, cs.shape)
self.assertEqual(X.shape + (7, ), ds.shape)
dd = cs[:, :, 1:] - cs[:, :, :-1]
assert_almost_equal(dd / 100, ds[:, :, 1:] / 100, decimal=4)
def test_fit_sirc_simulate(self):
model = CovidSIRDc()
X, y = model.iterate2array(derivatives=True)
epi = EpidemicRegressor('sirc', max_iter=10)
epi.fit(X, y)
sim = epi.simulate(X[4:6])
self.assertEqual(sim.shape, (2, 7, 4))
self.assertEqual(X[4].tolist(), sim[0, 0].tolist())
self.assertEqual(X[5].tolist(), sim[1, 0].tolist())
sim = epi.simulate(numpy.array([[10000., 0, 0, 0]]))
self.assertEqual(sim[0].tolist(), sim[-1].tolist())
def test_fit_sir(self):
model = CovidSIRD()
X, y = model.iterate2array(derivatives=True)
with self.assertRaises(ValueError):
EpidemicRegressor('sir2')
with self.assertRaises(TypeError):
EpidemicRegressor('sirc', init='r')
epi = EpidemicRegressor(max_iter=10)
with self.assertRaises(RuntimeError):
epi.score(X, y)
with self.assertRaises(RuntimeError):
epi.predict(X)
epi.fit(X, y)
loss = epi.score(X, y)
self.assertGreater(loss, 0)
l1 = epi.score(X, y, 'l1')
self.assertGreater(l1, 0)
with self.assertRaises(ValueError):
epi.score(X, y, 'l2')
pars = epi.coef_
self.assertIsInstance(pars, dict)
epi2 = EpidemicRegressor(max_iter=10, init=epi)
self.assertEqual(epi.coef_, epi2.coef_)
######################################
# Visualisation
df.plot(title="Simulation SIRD")
###########################################
# Estimation
# ++++++++++
#
# Le module implémente la class :class:`EpidemicRegressor
# <aftercovid.models.EpidemicRegressor>` qui réplique
# l'API de :epkg:`scikit-learn`.
m = EpidemicRegressor('SIRD',
verbose=True,
learning_rate_init=1e-3,
max_iter=10,
early_th=1)
m.fit(X, y)
pprint(m.model_.P)
###############################################
# La réaction de la population n'est pas constante
# tout au long de l'épidémie. Il est possible qu'elle
# change de comportement tout au long de la propagation.
# On estime alors les coefficients du modèle sur une
# fenêtre glissante.
def find_best_model(Xt, yt, lrs, th):
best_est, best_loss = None, None