mirror of
https://github.com/saymrwulf/prophet.git
synced 2026-07-03 03:59:00 +00:00
add implementation for constant trend in Python (#1466)
* add implementation for constant trend * force k and delta params to be 0s * add tests and fix n_changepoints, changepoints_t to 0 * Add test for cv with constant trend * Add docs and test for checking invalid input * make changes to stan * add transformed params block in stan and output flat trend vector * correct syntax * transformed params syntax * Fix test and port changes to win stan file * add test for flat trend function * Add separate function for flat trend init * fix test
This commit is contained in:
Ryan Nazareth
GitHub
parent
cb4f1dcc59
commit
ac59b44ca3
6 changed files with 212 additions and 56 deletions
File diff suppressed because one or more lines are too long
|
|
@ -155,9 +155,9 @@ class Prophet(object):
|
|||
|
||||
def validate_inputs(self):
|
||||
"""Validates the inputs to Prophet."""
|
||||
if self.growth not in ('linear', 'logistic'):
|
||||
if self.growth not in ('linear', 'logistic', 'flat'):
|
||||
raise ValueError(
|
||||
'Parameter "growth" should be "linear" or "logistic".')
|
||||
'Parameter "growth" should be "linear", "logistic" or "flat".')
|
||||
if ((self.changepoint_range < 0) or (self.changepoint_range > 1)):
|
||||
raise ValueError('Parameter "changepoint_range" must be in [0, 1]')
|
||||
if self.holidays is not None:
|
||||
|
|
@ -1049,6 +1049,27 @@ class Prophet(object):
|
|||
k = (L0 - L1) / T
|
||||
return (k, m)
|
||||
|
||||
@staticmethod
|
||||
def flat_growth_init(df):
|
||||
"""Initialize flat growth.
|
||||
|
||||
Provides a strong initialization for flat growth. Sets the growth to 0
|
||||
and offset parameter as mean of history y_scaled values.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
df: pd.DataFrame with columns ds (date), y_scaled (scaled time series),
|
||||
and t (scaled time).
|
||||
|
||||
Returns
|
||||
-------
|
||||
A tuple (k, m) with the rate (k) and offset (m) of the linear growth
|
||||
function.
|
||||
"""
|
||||
k = 0
|
||||
m = df['y_scaled'].mean()
|
||||
return k, m
|
||||
|
||||
def fit(self, df, **kwargs):
|
||||
"""Fit the Prophet model.
|
||||
|
||||
|
|
@ -1098,6 +1119,8 @@ class Prophet(object):
|
|||
|
||||
self.set_changepoints()
|
||||
|
||||
trend_indicator = {'linear': 0, 'logistic': 1, 'flat': 2}
|
||||
|
||||
dat = {
|
||||
'T': history.shape[0],
|
||||
'K': seasonal_features.shape[1],
|
||||
|
|
@ -1108,7 +1131,7 @@ class Prophet(object):
|
|||
'X': seasonal_features,
|
||||
'sigmas': prior_scales,
|
||||
'tau': self.changepoint_prior_scale,
|
||||
'trend_indicator': int(self.growth == 'logistic'),
|
||||
'trend_indicator': trend_indicator[self.growth],
|
||||
's_a': component_cols['additive_terms'],
|
||||
's_m': component_cols['multiplicative_terms'],
|
||||
}
|
||||
|
|
@ -1116,6 +1139,9 @@ class Prophet(object):
|
|||
if self.growth == 'linear':
|
||||
dat['cap'] = np.zeros(self.history.shape[0])
|
||||
kinit = self.linear_growth_init(history)
|
||||
elif self.growth == 'flat':
|
||||
dat['cap'] = np.zeros(self.history.shape[0])
|
||||
kinit = self.flat_growth_init(history)
|
||||
else:
|
||||
dat['cap'] = history['cap_scaled']
|
||||
kinit = self.logistic_growth_init(history)
|
||||
|
|
@ -1128,9 +1154,8 @@ class Prophet(object):
|
|||
'sigma_obs': 1,
|
||||
}
|
||||
|
||||
if (history['y'].min() == history['y'].max()
|
||||
and self.growth == 'linear'):
|
||||
# Nothing to fit.
|
||||
if history['y'].min() == history['y'].max() and \
|
||||
(self.growth == 'linear' or self.growth == 'flat'):
|
||||
self.params = stan_init
|
||||
self.params['sigma_obs'] = 1e-9
|
||||
for par in self.params:
|
||||
|
|
@ -1255,6 +1280,22 @@ class Prophet(object):
|
|||
m_t[indx] += gammas[s]
|
||||
return cap / (1 + np.exp(-k_t * (t - m_t)))
|
||||
|
||||
@staticmethod
|
||||
def flat_trend(t, m):
|
||||
"""Evaluate the flat trend function.
|
||||
|
||||
Parameters
|
||||
----------
|
||||
t: np.array of times on which the function is evaluated.
|
||||
m: Float initial offset.
|
||||
|
||||
Returns
|
||||
-------
|
||||
Vector y(t).
|
||||
"""
|
||||
m_t = m * np.ones_like(t)
|
||||
return m_t
|
||||
|
||||
def predict_trend(self, df):
|
||||
"""Predict trend using the prophet model.
|
||||
|
||||
|
|
@ -1273,10 +1314,13 @@ class Prophet(object):
|
|||
t = np.array(df['t'])
|
||||
if self.growth == 'linear':
|
||||
trend = self.piecewise_linear(t, deltas, k, m, self.changepoints_t)
|
||||
else:
|
||||
elif self.growth == 'logistic':
|
||||
cap = df['cap_scaled']
|
||||
trend = self.piecewise_logistic(
|
||||
t, cap, deltas, k, m, self.changepoints_t)
|
||||
elif self.growth == 'flat':
|
||||
# constant trend
|
||||
trend = self.flat_trend(t, m)
|
||||
|
||||
return trend * self.y_scale + df['floor']
|
||||
|
||||
|
|
@ -1470,10 +1514,12 @@ class Prophet(object):
|
|||
|
||||
if self.growth == 'linear':
|
||||
trend = self.piecewise_linear(t, deltas, k, m, changepoint_ts)
|
||||
else:
|
||||
elif self.growth == 'logistic':
|
||||
cap = df['cap_scaled']
|
||||
trend = self.piecewise_logistic(t, cap, deltas, k, m,
|
||||
changepoint_ts)
|
||||
elif self.growth == 'flat':
|
||||
trend = self.flat_trend(t, m)
|
||||
|
||||
return trend * self.y_scale + df['floor']
|
||||
|
||||
|
|
|
|||
|
|
@ -106,18 +106,21 @@ class TestDiagnostics(TestCase):
|
|||
self.assertEqual(diagnostics.single_cutoff_forecast.call_count,
|
||||
forecasts)
|
||||
|
||||
|
||||
def test_cross_validation_logistic(self):
|
||||
df = self.__df.copy()
|
||||
df['cap'] = 40
|
||||
m = Prophet(growth='logistic').fit(df)
|
||||
df_cv = diagnostics.cross_validation(
|
||||
m, horizon='1 days', period='1 days', initial='140 days')
|
||||
self.assertEqual(len(np.unique(df_cv['cutoff'])), 2)
|
||||
self.assertTrue((df_cv['cutoff'] < df_cv['ds']).all())
|
||||
df_merged = pd.merge(df_cv, self.__df, 'left', on='ds')
|
||||
self.assertAlmostEqual(
|
||||
np.sum((df_merged['y_x'] - df_merged['y_y']) ** 2), 0.0)
|
||||
def test_cross_validation_logistic_or_flat_growth(self):
|
||||
params = (x for x in ['logistic', 'flat'])
|
||||
for growth in params:
|
||||
with self.subTest(i=growth):
|
||||
df = self.__df.copy()
|
||||
if growth == "logistic":
|
||||
df['cap'] = 40
|
||||
m = Prophet(growth=growth).fit(df)
|
||||
df_cv = diagnostics.cross_validation(
|
||||
m, horizon='1 days', period='1 days', initial='140 days')
|
||||
self.assertEqual(len(np.unique(df_cv['cutoff'])), 2)
|
||||
self.assertTrue((df_cv['cutoff'] < df_cv['ds']).all())
|
||||
df_merged = pd.merge(df_cv, self.__df, 'left', on='ds')
|
||||
self.assertAlmostEqual(
|
||||
np.sum((df_merged['y_x'] - df_merged['y_y']) ** 2), 0.0)
|
||||
|
||||
def test_cross_validation_extra_regressors(self):
|
||||
df = self.__df.copy()
|
||||
|
|
|
|||
|
|
@ -206,22 +206,41 @@ class TestProphet(TestCase):
|
|||
# Check for approximate shift invariance
|
||||
self.assertTrue((np.abs(fcst1['yhat'] - fcst2['yhat']) < 1).all())
|
||||
|
||||
def test_get_changepoints(self):
|
||||
m = Prophet()
|
||||
def test_flat_growth(self):
|
||||
m = Prophet(growth='flat')
|
||||
N = DATA.shape[0]
|
||||
history = DATA.head(N // 2).copy()
|
||||
m.fit(history)
|
||||
future = m.make_future_dataframe(N // 2, include_history=False)
|
||||
fcst = m.predict(future)
|
||||
m_ = m.params['m']
|
||||
k = m.params['k']
|
||||
self.assertEqual(k[0, 0], 0)
|
||||
self.assertEqual(fcst['trend'].unique(), m_*m.y_scale)
|
||||
self.assertEqual(np.round(m_[0,0]*m.y_scale), 26)
|
||||
|
||||
history = m.setup_dataframe(history, initialize_scales=True)
|
||||
m.history = history
|
||||
def test_invalid_growth_input(self):
|
||||
msg = 'Parameter "growth" should be "linear", ' \
|
||||
'"logistic" or "flat".'
|
||||
with self.assertRaisesRegex(ValueError, msg):
|
||||
Prophet(growth="constant")
|
||||
|
||||
m.set_changepoints()
|
||||
def test_get_changepoints(self):
|
||||
m = Prophet()
|
||||
N = DATA.shape[0]
|
||||
history = DATA.head(N // 2).copy()
|
||||
|
||||
cp = m.changepoints_t
|
||||
self.assertEqual(cp.shape[0], m.n_changepoints)
|
||||
self.assertEqual(len(cp.shape), 1)
|
||||
self.assertTrue(cp.min() > 0)
|
||||
cp_indx = int(np.ceil(0.8 * history.shape[0]))
|
||||
self.assertTrue(cp.max() <= history['t'].values[cp_indx])
|
||||
history = m.setup_dataframe(history, initialize_scales=True)
|
||||
m.history = history
|
||||
|
||||
m.set_changepoints()
|
||||
|
||||
cp = m.changepoints_t
|
||||
self.assertEqual(cp.shape[0], m.n_changepoints)
|
||||
self.assertEqual(len(cp.shape), 1)
|
||||
self.assertTrue(cp.min() > 0)
|
||||
cp_indx = int(np.ceil(0.8 * history.shape[0]))
|
||||
self.assertTrue(cp.max() <= history['t'].values[cp_indx])
|
||||
|
||||
def test_set_changepoint_range(self):
|
||||
m = Prophet(changepoint_range=0.4)
|
||||
|
|
@ -301,6 +320,10 @@ class TestProphet(TestCase):
|
|||
self.assertAlmostEqual(k, 1.507925, places=4)
|
||||
self.assertAlmostEqual(m, -0.08167497, places=4)
|
||||
|
||||
k,m = model.flat_growth_init(history)
|
||||
self.assertEqual(k, 0)
|
||||
self.assertAlmostEqual(m, 0.49335657, places=4)
|
||||
|
||||
def test_piecewise_linear(self):
|
||||
model = Prophet()
|
||||
|
||||
|
|
@ -342,6 +365,18 @@ class TestProphet(TestCase):
|
|||
y = model.piecewise_logistic(t, cap, deltas, k, m, changepoint_ts)
|
||||
self.assertAlmostEqual((y - y_true).sum(), 0.0, places=5)
|
||||
|
||||
def test_flat_trend(self):
|
||||
model = Prophet()
|
||||
t = np.arange(11)
|
||||
m = 0.5
|
||||
y = model.flat_trend(t, m)
|
||||
y_true = np.array([0.5]*11)
|
||||
self.assertEqual((y - y_true).sum(), 0)
|
||||
t = t[8:]
|
||||
y_true = y_true[8:]
|
||||
y = model.flat_trend(t, m)
|
||||
self.assertEqual((y - y_true).sum(), 0)
|
||||
|
||||
def test_holidays(self):
|
||||
holidays = pd.DataFrame({
|
||||
'ds': pd.to_datetime(['2016-12-25']),
|
||||
|
|
|
|||
|
|
@ -73,6 +73,15 @@ functions {
|
|||
) {
|
||||
return (k + A * delta) .* t + (m + A * (-t_change .* delta));
|
||||
}
|
||||
|
||||
// Flat trend function
|
||||
|
||||
vector flat_trend(
|
||||
real m,
|
||||
int T
|
||||
) {
|
||||
return rep_vector(m, T);
|
||||
}
|
||||
}
|
||||
|
||||
data {
|
||||
|
|
@ -86,7 +95,7 @@ data {
|
|||
matrix[T,K] X; // Regressors
|
||||
vector[K] sigmas; // Scale on seasonality prior
|
||||
real<lower=0> tau; // Scale on changepoints prior
|
||||
int trend_indicator; // 0 for linear, 1 for logistic
|
||||
int trend_indicator; // 0 for linear, 1 for logistic, 2 for flat
|
||||
vector[K] s_a; // Indicator of additive features
|
||||
vector[K] s_m; // Indicator of multiplicative features
|
||||
}
|
||||
|
|
@ -104,6 +113,17 @@ parameters {
|
|||
vector[K] beta; // Regressor coefficients
|
||||
}
|
||||
|
||||
transformed parameters {
|
||||
vector[T] trend;
|
||||
if (trend_indicator == 0) {
|
||||
trend = linear_trend(k, m, delta, t, A, t_change);
|
||||
} else if (trend_indicator == 1) {
|
||||
trend = logistic_trend(k, m, delta, t, cap, A, t_change, S);
|
||||
} else if (trend_indicator == 2) {
|
||||
trend = flat_trend(m, T);
|
||||
}
|
||||
}
|
||||
|
||||
model {
|
||||
//priors
|
||||
k ~ normal(0, 5);
|
||||
|
|
@ -113,19 +133,10 @@ model {
|
|||
beta ~ normal(0, sigmas);
|
||||
|
||||
// Likelihood
|
||||
if (trend_indicator == 0) {
|
||||
y ~ normal(
|
||||
linear_trend(k, m, delta, t, A, t_change)
|
||||
.* (1 + X * (beta .* s_m))
|
||||
+ X * (beta .* s_a),
|
||||
sigma_obs
|
||||
);
|
||||
} else if (trend_indicator == 1) {
|
||||
y ~ normal(
|
||||
logistic_trend(k, m, delta, t, cap, A, t_change, S)
|
||||
.* (1 + X * (beta .* s_m))
|
||||
+ X * (beta .* s_a),
|
||||
sigma_obs
|
||||
);
|
||||
}
|
||||
y ~ normal(
|
||||
trend
|
||||
.* (1 + X * (beta .* s_m))
|
||||
+ X * (beta .* s_a),
|
||||
sigma_obs
|
||||
);
|
||||
}
|
||||
|
|
|
|||
|
|
@ -94,6 +94,17 @@ functions {
|
|||
}
|
||||
return Y;
|
||||
}
|
||||
|
||||
// Flat trend function
|
||||
|
||||
real[] flat_trend(
|
||||
real m,
|
||||
int T
|
||||
) {
|
||||
return rep_array(m, T);
|
||||
}
|
||||
|
||||
|
||||
}
|
||||
|
||||
data {
|
||||
|
|
@ -107,7 +118,7 @@ data {
|
|||
real X[T,K]; // Regressors
|
||||
vector[K] sigmas; // Scale on seasonality prior
|
||||
real<lower=0> tau; // Scale on changepoints prior
|
||||
int trend_indicator; // 0 for linear, 1 for logistic
|
||||
int trend_indicator; // 0 for linear, 1 for logistic, 2 for flat
|
||||
real s_a[K]; // Indicator of additive features
|
||||
real s_m[K]; // Indicator of multiplicative features
|
||||
}
|
||||
|
|
@ -135,6 +146,8 @@ transformed parameters {
|
|||
trend = linear_trend(k, m, delta, t, A, t_change, S, T);
|
||||
} else if (trend_indicator == 1) {
|
||||
trend = logistic_trend(k, m, delta, t, cap, A, t_change, S, T);
|
||||
} else if (trend_indicator == 2){
|
||||
trend = flat_trend(m, T);
|
||||
}
|
||||
|
||||
for (i in 1:K) {
|
||||
|
|
|
|||
Loading…
Reference in a new issue