def test_untrained_models(self):
model = NaiveDrift()
_ = NaiveEnsembleModel([model])
# trained models should raise error
model.fit(self.series1)
with self.assertRaises(ValueError):
NaiveEnsembleModel([model])
def test_input_models_local_models(self):
with self.assertRaises(ValueError):
NaiveEnsembleModel([])
with self.assertRaises(ValueError):
NaiveEnsembleModel(
[NaiveDrift, NaiveSeasonal, Theta, ExponentialSmoothing])
with self.assertRaises(ValueError):
NaiveEnsembleModel(
[NaiveDrift(), NaiveSeasonal,
Theta(),
ExponentialSmoothing()])
NaiveEnsembleModel(
[NaiveDrift(),
NaiveSeasonal(),
Theta(),
ExponentialSmoothing()])
def test_fit_univar_ts_with_covariates_for_local_models(self):
naive = NaiveEnsembleModel(
[NaiveDrift(),
NaiveSeasonal(),
Theta(),
ExponentialSmoothing()])
with self.assertRaises(ValueError):
naive.fit(self.series1, self.series2)
def test_backtest_forecasting(self):
linear_series = lt(length=50)
linear_series_multi = linear_series.stack(linear_series)
# univariate model + univariate series
score = NaiveDrift().backtest(linear_series,
start=pd.Timestamp('20000201'),
forecast_horizon=3,
metric=r2_score)
self.assertEqual(score, 1.0)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
start=pd.Timestamp('20000217'),
forecast_horizon=3)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
start=pd.Timestamp('20000217'),
forecast_horizon=3,
overlap_end=False)
NaiveDrift().backtest(linear_series,
start=pd.Timestamp('20000216'),
forecast_horizon=3)
NaiveDrift().backtest(linear_series,
start=pd.Timestamp('20000217'),
forecast_horizon=3,
overlap_end=True)
# Using forecast_horizon default value
NaiveDrift().backtest(linear_series, start=pd.Timestamp('20000216'))
NaiveDrift().backtest(linear_series,
start=pd.Timestamp('20000217'),
overlap_end=True)
# Using an int or float value for start
NaiveDrift().backtest(linear_series, start=30)
NaiveDrift().backtest(linear_series, start=0.7, overlap_end=True)
# Using invalid start and/or forecast_horizon values
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
start=0.7,
forecast_horizon=-1)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
start=-0.7,
forecast_horizon=1)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, start=100)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, start=1.2)
with self.assertRaises(TypeError):
NaiveDrift().backtest(linear_series, start='wrong type')
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
start=49,
forecast_horizon=2,
overlap_end=False)
# univariate model + multivariate series
with self.assertRaises(AssertionError):
NaiveDrift().backtest(linear_series_multi,
start=pd.Timestamp('20000201'),
forecast_horizon=3)
# multivariate model + univariate series
if TORCH_AVAILABLE:
tcn_model = TCNModel(input_chunk_length=12,
output_chunk_length=1,
batch_size=1,
n_epochs=1)
pred = tcn_model.historical_forecasts(
linear_series,
start=pd.Timestamp('20000125'),
forecast_horizon=3,
verbose=False,
last_points_only=True)
self.assertEqual(pred.width, 1)
self.assertEqual(pred.end_time(), linear_series.end_time())
# multivariate model + multivariate series
with self.assertRaises(ValueError):
tcn_model.backtest(linear_series_multi,
start=pd.Timestamp('20000125'),
forecast_horizon=3,
verbose=False)
tcn_model = TCNModel(input_chunk_length=12,
output_chunk_length=3,
batch_size=1,
n_epochs=1)
pred = tcn_model.historical_forecasts(
linear_series_multi,
start=pd.Timestamp('20000125'),
forecast_horizon=3,
verbose=False,
last_points_only=True)
self.assertEqual(pred.width, 2)
self.assertEqual(pred.end_time(), linear_series.end_time())
smape_all = []
m = int(info_dataset.Frequency[cat[0] + "1"])
for train, test in _build_tqdm_iterator(zip(ts_train, ts_test),
verbose=True):
train_des = train
seasonOut = 1
if m > 1:
if check_seasonality(train, m=m, max_lag=2 * m):
_, season = extract_trend_and_seasonality(
train, m, model=ModelMode.MULTIPLICATIVE)
train_des = remove_from_series(
train, season, model=ModelMode.MULTIPLICATIVE)
seasonOut = season[-m:].shift(m)
seasonOut = seasonOut.append_values(seasonOut.values())
seasonOut = seasonOut[:len(test)]
naive = NaiveDrift()
naive2 = NaiveSeasonal(K=1)
naiveSeason = NaiveSeasonal(K=m)
ses = ExponentialSmoothing(trend=None,
seasonal=None,
seasonal_periods=m)
holt = ExponentialSmoothing(seasonal=None,
damped=False,
trend='additive',
seasonal_periods=m)
damp = ExponentialSmoothing(seasonal=None,
damped=True,
trend='additive',
seasonal_periods=m)
naive.fit(train)
naive2.fit(train_des)
def test_backtest_forecasting(self):
linear_series = lt(length=50)
linear_series_int = TimeSeries.from_values(linear_series.values())
linear_series_multi = linear_series.stack(linear_series)
# univariate model + univariate series
score = NaiveDrift().backtest(
linear_series,
start=pd.Timestamp("20000201"),
forecast_horizon=3,
metric=r2_score,
)
self.assertEqual(score, 1.0)
# very large train length should not affect the backtest
score = NaiveDrift().backtest(
linear_series,
train_length=10000,
start=pd.Timestamp("20000201"),
forecast_horizon=3,
metric=r2_score,
)
self.assertEqual(score, 1.0)
# window of size 2 is too small for naive drift
with self.assertRaises(ValueError):
NaiveDrift().backtest(
linear_series,
train_length=2,
start=pd.Timestamp("20000201"),
forecast_horizon=3,
metric=r2_score,
)
# test that it also works for time series that are not Datetime-indexed
score = NaiveDrift().backtest(
linear_series_int, start=0.7, forecast_horizon=3, metric=r2_score
)
self.assertEqual(score, 1.0)
with self.assertRaises(ValueError):
NaiveDrift().backtest(
linear_series,
start=pd.Timestamp("20000217"),
forecast_horizon=3,
overlap_end=False,
)
NaiveDrift().backtest(
linear_series, start=pd.Timestamp("20000216"), forecast_horizon=3
)
NaiveDrift().backtest(
linear_series,
start=pd.Timestamp("20000217"),
forecast_horizon=3,
overlap_end=True,
)
# Using forecast_horizon default value
NaiveDrift().backtest(linear_series, start=pd.Timestamp("20000216"))
NaiveDrift().backtest(
linear_series, start=pd.Timestamp("20000217"), overlap_end=True
)
# Using an int or float value for start
NaiveDrift().backtest(linear_series, start=30)
NaiveDrift().backtest(linear_series, start=0.7, overlap_end=True)
# Set custom train window length
NaiveDrift().backtest(linear_series, train_length=10, start=30)
# Using invalid start and/or forecast_horizon values
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, start=0.7, forecast_horizon=-1)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, start=-0.7, forecast_horizon=1)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, start=100)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, start=1.2)
with self.assertRaises(TypeError):
NaiveDrift().backtest(linear_series, start="wrong type")
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, train_length=0, start=0.5)
with self.assertRaises(TypeError):
NaiveDrift().backtest(linear_series, train_length=1.2, start=0.5)
with self.assertRaises(TypeError):
NaiveDrift().backtest(linear_series, train_length="wrong type", start=0.5)
with self.assertRaises(ValueError):
NaiveDrift().backtest(
linear_series, start=49, forecast_horizon=2, overlap_end=False
)
# univariate model + multivariate series
with self.assertRaises(AssertionError):
NaiveDrift().backtest(
linear_series_multi, start=pd.Timestamp("20000201"), forecast_horizon=3
)
# multivariate model + univariate series
if TORCH_AVAILABLE:
tcn_model = TCNModel(
input_chunk_length=12, output_chunk_length=1, batch_size=1, n_epochs=1
)
pred = tcn_model.historical_forecasts(
linear_series,
start=pd.Timestamp("20000125"),
forecast_horizon=3,
verbose=False,
last_points_only=True,
)
self.assertEqual(pred.width, 1)
self.assertEqual(pred.end_time(), linear_series.end_time())
# multivariate model + multivariate series
with self.assertRaises(ValueError):
tcn_model.backtest(
linear_series_multi,
start=pd.Timestamp("20000125"),
forecast_horizon=3,
verbose=False,
)
tcn_model = TCNModel(
input_chunk_length=12, output_chunk_length=3, batch_size=1, n_epochs=1
)
pred = tcn_model.historical_forecasts(
linear_series_multi,
start=pd.Timestamp("20000125"),
forecast_horizon=3,
verbose=False,
last_points_only=True,
)
self.assertEqual(pred.width, 2)
self.assertEqual(pred.end_time(), linear_series.end_time())
def test_backtest_forecasting(self):
linear_series = lt(length=50)
linear_series_multi = linear_series.stack(linear_series)
# univariate model + univariate series
pred = NaiveDrift().backtest(linear_series, None,
pd.Timestamp('20000201'), 3)
self.assertEqual(r2_score(pred, linear_series), 1.0)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
None,
start=pd.Timestamp('20000217'),
forecast_horizon=3)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
None,
start=pd.Timestamp('20000217'),
forecast_horizon=3,
trim_to_series=True)
NaiveDrift().backtest(linear_series,
None,
start=pd.Timestamp('20000216'),
forecast_horizon=3)
NaiveDrift().backtest(linear_series,
None,
pd.Timestamp('20000217'),
forecast_horizon=3,
trim_to_series=False)
# Using forecast_horizon default value
NaiveDrift().backtest(linear_series,
None,
start=pd.Timestamp('20000216'))
NaiveDrift().backtest(linear_series,
None,
pd.Timestamp('20000217'),
trim_to_series=False)
# Using an int or float value for start
NaiveDrift().backtest(linear_series, None, start=30)
NaiveDrift().backtest(linear_series,
None,
start=0.7,
trim_to_series=False)
# Using invalid start and/or forecast_horizon values
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
None,
start=0.7,
forecast_horizon=-1)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, None, 0.7, -1)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, None, start=100)
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series, None, start=1.2)
with self.assertRaises(TypeError):
NaiveDrift().backtest(linear_series, None, start='wrong type')
with self.assertRaises(ValueError):
NaiveDrift().backtest(linear_series,
None,
start=49,
forecast_horizon=2,
trim_to_series=True)
# univariate model + multivariate series
with self.assertRaises(AssertionError):
NaiveDrift().backtest(linear_series_multi, None,
pd.Timestamp('20000201'), 3)
# multivariate model + univariate series
if TORCH_AVAILABLE:
tcn_model = TCNModel(batch_size=1, n_epochs=1)
pred = tcn_model.backtest(linear_series,
None,
pd.Timestamp('20000125'),
3,
verbose=False)
self.assertEqual(pred.width, 1)
# multivariate model + multivariate series
with self.assertRaises(ValueError):
tcn_model.backtest(linear_series_multi,
None,
pd.Timestamp('20000125'),
3,
verbose=False)
tcn_model = TCNModel(batch_size=1,
n_epochs=1,
input_size=2,
output_length=3)
with self.assertRaises(ValueError):
tcn_model.backtest(linear_series_multi,
None,
pd.Timestamp('20000125'),
3,
verbose=False,
use_full_output_length=False)
pred = tcn_model.backtest(linear_series_multi,
linear_series_multi[['0']],
pd.Timestamp('20000125'),
1,
verbose=False,
use_full_output_length=True)
self.assertEqual(pred.width, 1)
pred = tcn_model.backtest(linear_series_multi,
linear_series_multi[['1']],
pd.Timestamp('20000125'),
3,
verbose=False,
use_full_output_length=True)
self.assertEqual(pred.width, 1)
tcn_model = TCNModel(batch_size=1,
n_epochs=1,
input_size=2,
output_length=3,
output_size=2)
pred = tcn_model.backtest(linear_series_multi,
linear_series_multi,
pd.Timestamp('20000125'),
3,
verbose=False,
use_full_output_length=True)
self.assertEqual(pred.width, 2)
weight = st.number_input("Today's weight:",
min_value=50.0,
max_value=750.0,
format="%4.1f",
step=.1)
st.write("Echoing weight: ", weight)
sqlStoreWeight(todaysDate, weight)
#weightsDF = pd.DataFrame({'date':['7-17-2020','7-18-2020','7-19-2020','7-20-2020','7-21-2020','7-22-2020'],
# 'weight':[170.8, 170.4, 170.8, 170.6, 170.6, 745]})
weightsDF = sqlGetWeightsDF() # Would passing todaysEXPTDate here be helpful?
if weightsDF.date.size > 30: # Remember, size is rows*cols, so just get the size of a series or the index of the DF
glModel = AutoARIMA()
else:
glModel = NaiveDrift()
series = TimeSeries.from_dataframe(weightsDF,
time_col='date',
value_cols='weight')
train, val = series.split_after(pd.Timestamp(todaysDate -
dt.timedelta(days=3)))
trainTheModel(train)
next3daysDF = predict_next(4)
print(next3daysDF[0])
plotYmax = weightsDF['weight'].max() + 0.5
plotYmin = weightsDF['weight'].min() - 0.5
def get_local_models(self):
return [NaiveDrift(), NaiveSeasonal(5), NaiveSeasonal(10)]
def test_input_models_mixed(self):
with self.assertRaises(ValueError):
NaiveEnsembleModel([NaiveDrift(), Theta(), RNNModel(12)])