def test_inference():
# https://github.com/pandas-dev/pandas/blob/master/pandas/tseries/frequencies.py
expected = {'S':[RECENT_SECONDS + j for j in range(200)],
'5S':[RECENT_SECONDS + 5 * j for j in range(200)],
'T':[RECENT_SECONDS + 60 * j for j in range(200)],
'5T':[RECENT_SECONDS + 5 * 60 * j for j in range(200)]} # 5 minutes
for expected_freq, t in expected.items():
freq = infer_freq_from_epoch(t)
assert freq==expected_freq
def prophet_fit_and_predict_with_exog_and_advance_vars(
y: [[float]],
k: int,
t: [float],
a: [[float]],
model_params: dict = None) -> Tuple[List, List, Any, Any]:
""" Simpler wrapper for testing - univariate w/ advance vars w/ supplied times and future times """
assert len(t) == len(y) + k
assert len(a) == len(y) + k
assert isinstance(y[0], List)
a_cols = ['a' + str(i) for i in range(len(a[0]))]
df = pd.DataFrame(columns=a_cols, data=a[:-k])
Y = transpose(y)
df['y'] = Y[0]
n_exog = len(y[0]) - 1
y_cols = ['y'] + ['y' + str(i) for i in range(1, len(y[0]))]
for i in range(1, n_exog + 1):
df['y' + str(i)] = Y[i][:len(y)]
dt = epoch_to_naive_datetime(t)
df['ds'] = dt[:len(y)]
kwargs_used = dict([(k, v) for k, v in PROPHET_MODEL.items()])
if model_params:
kwargs_used.update(model_params)
m = Prophet(**kwargs_used)
for a_col in a_cols:
m.add_regressor(name=a_col)
for y_col in y_cols[1:]:
m.add_regressor(name=y_col)
with no_stdout_stderr():
m.fit(df)
freq = infer_freq_from_epoch(t)
future = m.make_future_dataframe(periods=k, freq=freq)
future['ds'] = dt
full_a_data = transpose(a)
for a_col, a_vals in zip(a_cols, full_a_data):
future[a_col] = a_vals
for i in range(1, n_exog + 1): # Just bring forward
future['y' + str(i)] = Y[i] + [Y[i][-1]] * k
forecast = m.predict(future)
x = forecast['yhat'].values[-k:] # Use m.plot(forecast) to take a peak
x_std = [
u - l for u, l in zip(forecast['yhat_upper'].values,
forecast['yhat_lower'].values)
]
return x, x_std, forecast, m
def prophet_fit_and_predict_with_time_and_advance_time(y: [float], k: int, t: [float], model_params: dict = None) -> \
Tuple[List, List, Any, Any]:
""" Simpler wrapper for testing - univariate only w/ supplied times and future times """
assert len(t) == len(y) + k
df = pd.DataFrame(columns=['y'], data=y)
dt = epoch_to_naive_datetime(t)
df['ds'] = dt[:len(y)]
kwargs_used = dict([(k, v) for k, v in PROPHET_MODEL.items()])
if model_params:
kwargs_used.update(model_params)
m = Prophet(**kwargs_used)
with no_stdout_stderr():
m.fit(df)
freq = infer_freq_from_epoch(t)
future = m.make_future_dataframe(periods=k, freq=freq)
future['ds'] = dt
forecast = m.predict(future)
x = forecast['yhat'].values[-k:] # Use m.plot(forecast) to take a peak
x_std = [
u - l for u, l in zip(forecast['yhat_upper'].values,
forecast['yhat_lower'].values)
]
return x, x_std, forecast, m
def prophet_iskater_factory(y: [[float]],
k: int,
a: List = None,
t: List = None,
e=None,
freq: str = None,
n_max=1000,
recursive: bool = False,
model_params: dict = None,
return_forecast=True):
"""
:param y: A list of observations, each a vector.
:param k: Number of steps ahead to predict
:param a: Known in advance observations - should be k more of these than y's
:param t: Epoch times of observations y. If len(t)=len(y)+k the last k are interpreted as future times.
:param freq: 'D', '5T' etc, see https://github.com/pandas-dev/pandas/blob/master/pandas/tseries/frequencies.py
:param n_max: Maximum number of observations to use, should you wish to prevent prophet from slowing down
:param recursive If True, exogenous variables y[1], y[2],... will be predicted forward in time
(obviously this adds to computation time)
:returns: x k-vector of predictions
x_std k-vector of standard deviations
forecast full forecast dataframe, familiar to users of fbprophet
"""
if a:
assert len(a) == len(y) + k
if isinstance(y[0], float):
y = [wrap(yj) for yj in y]
# Conversion of epoch times to UTC datetime
# User must supply times, len(y) or len(y)+k, or a valid frequency str
if t is None:
if freq is None or not freq:
freq = PROPHET_META['freq'] # Just assume away ...
else:
assert is_valid_freq(
freq), 'Freq ' + str(freq) + ' is not a valid frequency'
dt = pd.date_range(start=EPOCH, periods=len(y), freq=freq) # UTC
else:
freq = infer_freq_from_epoch(t)
dt = epoch_to_naive_datetime(t)
if len(dt) == len(y) + k:
ta = dt
dt = dt[:len(y)]
else:
assert len(dt) == len(
y), 'Time vector t should be len(y) or len(y)+k'
ta = None
# Truncate history so that prophet doesn't take forever to fit
y_shorter = y[-n_max:]
a_shorter = a[-(n_max + k):] if a is not None else [] # may be empty
dt_shorter = dt[-n_max:]
# Massage data into Prophet friendly dataframe with columns y, y1, ..., yk, a0,...aj
y_cols = [
'y' + str(i) if i > 0 else 'y' for i in range(len(y_shorter[-1]))
]
if a:
a_cols = ['a' + str(i) for i in range(len(a_shorter[-1]))]
data = [
list(yi) + list(ai)
for yi, ai in zip(y_shorter, a_shorter[:-k])
]
df = pd.DataFrame(columns=y_cols + a_cols, data=data)
else:
data = [list(yi) for yi in y_shorter]
df = pd.DataFrame(columns=y_cols, data=data)
df['ds'] = dt_shorter
# Instantiate Prophet model, ensure defaults are what we think they are
kwargs_used = dict([(k, v) for k, v in PROPHET_MODEL.items()])
if model_params:
kwargs_used.update(model_params)
m = Prophet(**kwargs_used)
# Add regressors
for y_col in y_cols[1:]:
m.add_regressor(name=y_col)
if a:
for a_col in a_cols:
m.add_regressor(name=a_col)
# Fit the model every invocation ... there isn't any other way
with no_stdout_stderr():
m.fit(df)
# Make future dataframe, adding known-in-advance variables
future = m.make_future_dataframe(periods=k, freq=freq)
if a:
for j, a_col in enumerate(a_cols):
future[a_col] = [ai[j] for ai in a_shorter] # Known in advance
if ta is not None:
future['ds'] = ta # override with user supplied future times
# Next, we wish to add contemporaneously observed variables
#
# This is somewhat problematic, for how should we bring exogenously observed variables forward?
# The simplest answer is, don't use them - only supply 1-vector y observations
# prophet implicitly assumes all exogenous are known, which is a pretty big shortcoming.
#
# However, if we are trying to support y[1:], ...
# - It seems consistent to use prophet to predict these forward,
# - It also seems likely that this will lead to over-fitting.
# I'm open to ideas here. Perhaps perform some hackery could effect attenuation of the coefficients
# assigned to y[1],... such as jiggling past observations. For now we use prophet on each
# one individually, feeding them the known in advance 'a' variables.
n_exog = len(y[0]) - 1
if n_exog > 0:
for j, y_col in enumerate(y_cols):
if j > 0:
yj = [yi[j] for yi in y_shorter]
if recursive:
yj_hat, yj_hat_std, yj_forecast, yj_m = prophet_iskater_factory(
y=yj,
k=k,
a=a_shorter,
freq=freq,
n_max=n_max,
recursive=False)
else:
yj_hat = [yj[-1]] * k
future[y_col] = yj + list(yj_hat)
# Call the prediction function
forecast = m.predict(future)
x = list(forecast['yhat'].values[-k:]
) # Use m.plot(forecast) to take a peak
# Interpret confidence level difference as scale to be returned. TODO: set alpha properly so this really is 1-std
x_std = list([
u - l for u, l in zip(forecast['yhat_upper'].values[-k:],
forecast['yhat_lower'].values[-k:])
])
if return_forecast:
return x, x_std, forecast, m
else:
return x, x_std
