ODE parameter prediction
We will perform ODE parameter prediction for forecasting of the number of cases. We have two ways for prediction.
Time-series prediction withOUT indicators
Time-series prediction with indicators
The second one uses indicators, including OxCGRT indicators, the number of vaccinations.
Note:
The target (Y) of prediction is ODE parameter values, not the number of cases. ODE parameter values are more useful because ODE parameter values have physical meanings, including (non-dimensional) effective contact rate, and they are always in the range of (0, 1).
[1]:
from datetime import timedelta
import covsirphy as cs
from matplotlib import pyplot as plt
import numpy as np
cs.__version__
[1]:
'3.1.2.delta'
The target of prediction is estimated ODE parameter values. At we will prepare them as explained with tutorials. For example, we can use class method ODEScenario.auto_build(), specifying location name. “Baseline” scenario will be created automatically with downloaded datasets.
[2]:
snr = cs.ODEScenario.auto_build(geo="Japan", model=cs.SIRFModel)
# Show summary
snr.summary()
2025-11-22 at 12:39:32 | INFO |
<SIR-F Model: parameter estimation>
2025-11-22 at 12:39:32 | INFO | Running optimization with 4 CPUs...
2025-11-22 at 12:42:02 | INFO | Completed optimization. Total: 2 min 29 sec
[2]:
| Start | End | Rt | theta | kappa | rho | sigma | alpha1 [-] | 1/alpha2 [day] | 1/beta [day] | 1/gamma [day] | ODE | tau | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Scenario | Phase | |||||||||||||
| Baseline | 0th | 2020-02-23 | 2020-08-10 | 1.96 | 0.203229 | 0.000001 | 0.00121 | 0.000491 | 0.203 | 20281 | 14 | 34 | SIR-F Model | 24 |
| 1st | 2020-08-11 | 2020-11-16 | 0.88 | 0.003295 | 0.000006 | 0.001109 | 0.001255 | 0.003 | 2965 | 15 | 13 | SIR-F Model | 24 | |
| 2nd | 2020-11-17 | 2020-12-24 | 1.53 | 0.000049 | 0.000014 | 0.001334 | 0.00086 | 0.0 | 1163 | 12 | 19 | SIR-F Model | 24 | |
| 3rd | 2020-12-25 | 2021-01-16 | 1.62 | 0.001362 | 0.000013 | 0.001172 | 0.000708 | 0.001 | 1307 | 14 | 24 | SIR-F Model | 24 | |
| 4th | 2021-01-17 | 2021-02-10 | 0.73 | 0.000327 | 0.000016 | 0.000734 | 0.000991 | 0.0 | 1032 | 23 | 17 | SIR-F Model | 24 | |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | |
| 63rd | 2023-03-12 | 2023-03-27 | 1.68 | 0.001 | 0.000004 | 0.001027 | 0.000607 | 0.001 | 4042 | 16 | 27 | SIR-F Model | 24 | |
| 64th | 2023-03-28 | 2023-04-07 | 0.68 | 0.001059 | 0.000004 | 0.001151 | 0.001677 | 0.001 | 3819 | 14 | 10 | SIR-F Model | 24 | |
| 65th | 2023-04-08 | 2023-04-18 | 1.46 | 0.000136 | 0.000003 | 0.001388 | 0.000948 | 0.0 | 5210 | 12 | 18 | SIR-F Model | 24 | |
| 66th | 2023-04-19 | 2023-04-27 | 1.76 | 0.001616 | 0.000003 | 0.00156 | 0.000884 | 0.002 | 5947 | 11 | 19 | SIR-F Model | 24 | |
| 67th | 2023-04-28 | 2023-05-08 | 1.55 | 0.000196 | 0.000002 | 0.001536 | 0.000989 | 0.0 | 8752 | 11 | 17 | SIR-F Model | 24 |
68 rows × 13 columns
For demonstration, we will get the start date of future phases.
[3]:
future_start_date = snr.simulate(display=False).index.max() + timedelta(days=1)
future_start_date
[3]:
Timestamp('2023-05-09 00:00:00')
1. Time-series prediction withOUT indicators
This scenario “Predicted” does not use indicators, using AutoTS package: a time series package for Python designed for rapidly deploying high-accuracy forecasts at scale.
At first, create “Predicted” scenario, copying estimated ODE parameter values of “Baseline” scenario.
[4]:
snr.build_with_template(name="Predicted", template="Baseline");
Then, run ODEScenario().predict(days, name, **kwargs). We can apply keyword arguments of autots.AutoTS() except for forecast_length (always the same as days).
[5]:
snr.predict(days=30, name="Predicted");
Using 1 cpus for n_jobs.
Data frequency is: D, used frequency is: D
Model Number: 1 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "SinTrend"}, "transformation_params": {"0": {}, "1": {}}}
Model Number: 2 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 3 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "Round", "2": "Detrend"}, "transformation_params": {"0": {"lag_1": 7, "method": "Mean"}, "1": {"model": "middle", "decimals": 2, "on_transform": true, "on_inverse": false}, "2": {"model": "GLS"}}}
Model Number: 4 with model GLS in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
Model Number: 5 with model GLS in generation 0 of 1 with params {} and transformations {"fillna": "median", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "RobustScaler", "3": "Round", "4": "MaxAbsScaler"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}, "3": {"model": "middle", "decimals": 2, "on_transform": true, "on_inverse": true}, "4": {}}}
Model Number: 6 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "bkfilter", "1": "SinTrend", "2": "Detrend", "3": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {}, "2": {"model": "Linear"}, "3": {}}}
Model Number: 7 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PositiveShift", "1": "SinTrend", "2": "bkfilter"}, "transformation_params": {"0": {}, "1": {}, "2": {}}}
Model Number: 8 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "SinTrend"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}}}
Model Number: 9 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 10 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 2, "lag_2": 7} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SinTrend", "1": "Round", "2": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {"model": "middle", "decimals": 2, "on_transform": false, "on_inverse": true}, "2": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6580, in _fit_one
df = self.transformers[i].fit_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/utils/_set_output.py", line 319, in wrapped
data_to_wrap = f(self, X, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/base.py", line 1389, in wrapper
return fit_method(estimator, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py", line 3320, in fit_transform
return self._fit(X, y, force_transform=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py", line 3352, in _fit
self.lambdas_[i] = optim_function(col)
^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py", line 3530, in _yeo_johnson_optimize
return optimize.brent(_neg_log_likelihood, brack=(-2, 2))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2626, in brent
res = _minimize_scalar_brent(func, brack, args, **options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2663, in _minimize_scalar_brent
brent.optimize()
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2433, in optimize
xa, xb, xc, fa, fb, fc, funcalls = self.get_bracket_info()
^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2402, in get_bracket_info
xa, xb, xc, fa, fb, fc, funcalls = bracket(func, xa=brack[0],
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 3032, in bracket
raise e
scipy.optimize._optimize.BracketError: The algorithm terminated without finding a valid bracket. Consider trying different initial points.
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 922, in fit
df_train_transformed = self.transformer_object._fit(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer PowerTransformer failed on fit from params rolling_mean_24 {'0': {}, '1': {'model': 'middle', 'decimals': 2, 'on_transform': False, 'on_inverse': True}, '2': {}} with error BracketError('The algorithm terminated without finding a valid bracket. Consider trying different initial points.')
in model 10 in generation 0: SeasonalNaive
Model Number: 11 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 2, "lag_2": 1} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "QuantileTransformer", "2": "Detrend"}, "transformation_params": {"0": {"lag_1": 12, "method": "Median"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"model": "GLS"}}}
Model Number: 12 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 7, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "QuantileTransformer", "1": "ClipOutliers"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 13 with model ConstantNaive in generation 0 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer", "2": "SeasonalDifference"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 7, "method": "LastValue"}}}
/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py:3475: RuntimeWarning: overflow encountered in power
out[pos] = (np.power(x[pos] + 1, lmbda) - 1) / lmbda
Model Number: 14 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 28} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 15 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "sokalmichener", "include_differenced": true, "k": 20, "stride_size": 1, "regression_type": null} and transformations {"fillna": "zero", "transformations": {"0": null}, "transformation_params": {"0": {}}}
Model Number: 16 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer", "2": "RobustScaler"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 17 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 30} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 18 with model SeasonalityMotif in generation 0 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "mae", "k": 10, "datepart_method": "common_fourier"} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
Model Number: 19 with model SectionalMotif in generation 0 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "AlignLastValue", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"lag_1": 7, "method": "Median"}, "1": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}}}
Model Number: 20 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "simple_binarized", "changepoint_spacing": 360, "changepoint_distance_end": 360, "regression_type": null, "lambda_": 0.01} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "Round", "1": "AlignLastValue", "2": "HistoricValues", "3": "ClipOutliers", "4": "bkfilter"}, "transformation_params": {"0": {"decimals": -1, "on_transform": true, "on_inverse": true}, "1": {"rows": 7, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": true, "threshold": null, "threshold_method": "mean"}, "2": {"window": 10}, "3": {"method": "clip", "std_threshold": 3, "fillna": null}, "4": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3620, in fit
noise = np.random.normal(0, 0.001, size=Y_values)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "numpy/random/mtrand.pyx", line 1580, in numpy.random.mtrand.RandomState.normal
File "_common.pyx", line 657, in numpy.random._common.cont
ValueError: maximum supported dimension for an ndarray is currently 64, found 1141
in model 20 in generation 0: BasicLinearModel
Model Number: 21 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["dayofweek", [365.25, 14]], "changepoint_spacing": 90, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98} and transformations {"fillna": "piecewise_polynomial", "transformations": {"0": "AlignLastValue", "1": "IntermittentOccurrence", "2": "RobustScaler", "3": "Log"}, "transformation_params": {"0": {"rows": 1, "lag": 2, "method": "multiplicative", "strength": 0.9, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"center": "mean"}, "2": {}, "3": {}}}
Model Number: 22 with model SeasonalityMotif in generation 0 of 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
Model Number: 23 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "recurring", "changepoint_spacing": 28, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 0.01, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Slice", "1": "ClipOutliers", "2": "SeasonalDifference", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer", "5": "PositiveShift"}, "transformation_params": {"0": {"method": 0.2}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "5": {}}}
Model Number: 24 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "zero", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AnomalyRemoval"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": 30, "transform_dict": null}, "2": {"method": "zscore", "method_params": {"distribution": "uniform", "alpha": 0.14}, "fillna": "fake_date", "transform_dict": {"fillna": "quadratic", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 7, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}}}, "isolated_only": false, "on_inverse": false}}}
Model Number: 25 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastDiff", "1": "MaxAbsScaler", "2": "HolidayTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"rows": 28, "displacement_rows": 1, "quantile": 0.7, "decay_span": null}, "1": {}, "2": {"threshold": 0.8, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": false, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": true, "use_hindu_holidays": false, "anomaly_detector_params": {"method": "med_diff", "method_params": {"distribution": "beta", "alpha": 0.05}, "fillna": "ffill", "transform_dict": null, "isolated_only": true, "on_inverse": false}, "remove_excess_anomalies": false, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": [168, "hour"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {"rows": 1, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 26 with model AverageValueNaive in generation 0 of 1 with params {"method": "Midhinge", "window": null} and transformations {"fillna": "median", "transformations": {"0": "ScipyFilter", "1": "HPFilter", "2": "AlignLastValue", "3": "UpscaleDownscaleTransformer"}, "transformation_params": {"0": {"method": "savgol_filter", "method_args": {"window_length": 31, "polyorder": 1, "deriv": 0, "mode": "mirror"}}, "1": {"part": "trend", "lamb": 129600}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "3": {"mode": "upscale", "factor": 1, "down_method": "decimate", "fill_method": "linear"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2288, in TemplateWizard
template_result, best_smape = _eval_prediction_for_template(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1637, in _eval_prediction_for_template
model_error = df_forecast.evaluate(
^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/base.py", line 830, in evaluate
) = full_metric_evaluation(
^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/metrics.py", line 630, in full_metric_evaluation
full_errors = F - A
~~^~~
ValueError: operands could not be broadcast together with shapes (60,4) (30,4)
in model 26 in generation 0: AverageValueNaive
Model Number: 27 with model GLS in generation 0 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "ChangepointDetrend"}, "transformation_params": {"0": {"model": "Linear", "changepoint_spacing": 5040, "changepoint_distance_end": 28, "datepart_method": null}}}
Model Number: 28 with model SeasonalNaive in generation 0 of 1 with params {"method": "mean", "lag_1": 2, "lag_2": 60} and transformations {"fillna": "rolling_mean", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}}}
Model Number: 29 with model SeasonalityMotif in generation 0 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 30 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "yule", "include_differenced": true, "k": 3, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "EWMAFilter"}, "transformation_params": {"0": {"rows": null, "displacement_rows": 4, "quantile": 1.0, "decay_span": null}, "1": {"span": 12}}}, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "quadratic", "transformations": {"0": "AlignLastValue", "1": "PCA", "2": "QuantileTransformer", "3": "AlignLastValue", "4": "HPFilter", "5": "CenterSplit"}, "transformation_params": {"0": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "1": {"whiten": false, "n_components": 0.3}, "2": {"output_distribution": "uniform", "n_quantiles": 1000}, "3": {"rows": 168, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "max"}, "4": {"part": "trend", "lamb": 1600}, "5": {"fillna": "ffill", "center": "median"}}}
Model Number: 31 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill", "transformations": {"0": "LevelShiftTransformer", "1": "cffilter", "2": "MaxAbsScaler", "3": "SinTrend"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}, "1": {}, "2": {}, "3": {}}}
Model Number: 32 with model SeasonalityMotif in generation 0 of 1 with params {"window": 30, "point_method": "mean", "distance_metric": "mae", "k": 5, "datepart_method": "recurring", "independent": false} and transformations {"fillna": "median", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "RegressionFilter", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 5, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "2": {"sigma": 1, "rolling_window": 90, "run_order": "season_first", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 0.05}}, "datepart_method": "expanded_binarized", "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "holiday_params": null, "trend_method": "rolling_mean"}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 33 with model AverageValueNaive in generation 0 of 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 0.05}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
Model Number: 34 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 35 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 90, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "IntermittentOccurrence", "2": "Discretize", "3": "Discretize", "4": "AlignLastValue", "5": "RollingMeanTransformer"}, "transformation_params": {"0": {"rows": 28, "displacement_rows": 1, "quantile": 1.0, "decay_span": 365}, "1": {"center": "mean"}, "2": {"discretization": "upper", "n_bins": 10}, "3": {"discretization": "center", "n_bins": 20}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "5": {"fixed": true, "window": 3, "macro_micro": false, "center": true}}}
Model Number: 36 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 37 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "median", "distance_metric": "jaccard", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "rolling_mean", "transformations": {"0": "QuantileTransformer", "1": "AlignLastValue", "2": "StandardScaler", "3": "CumSumTransformer"}, "transformation_params": {"0": {"output_distribution": "normal", "n_quantiles": "fifth"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.7, "first_value_only": false, "threshold": 3, "threshold_method": "max"}, "2": {}, "3": {}}}
Model Number: 38 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 90, "changepoint_distance_end": null, "regression_type": null, "lambda_": 0.01, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "IntermittentOccurrence", "1": "AlignLastValue", "2": "AlignLastValue", "3": "RollingMeanTransformer", "4": "AlignLastValue", "5": "AlignLastValue"}, "transformation_params": {"0": {"center": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "3": {"fixed": false, "window": 12, "macro_micro": true, "center": false}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "5": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 39 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
Model Number: 40 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 41 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "akima", "transformations": {"0": "MinMaxScaler", "1": "SeasonalDifference", "2": "MinMaxScaler", "3": "EWMAFilter", "4": "LevelShiftTransformer", "5": "Detrend"}, "transformation_params": {"0": {}, "1": {"lag_1": 12, "method": "LastValue"}, "2": {}, "3": {"span": 10}, "4": {"window_size": 14, "alpha": 4.0, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "rolling_diff_5nn"}, "5": {"model": "Linear", "phi": 1, "window": null, "transform_dict": null}}}
Model Number: 42 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 70, "lag_2": null} and transformations {"fillna": "pchip", "transformations": {"0": "CenterLastValue", "1": "DifferencedTransformer", "2": "Detrend"}, "transformation_params": {"0": {"rows": 3}, "1": {"lag": 1, "fill": "bfill"}, "2": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}}}
Model Number: 43 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 180, "constant": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue", "4": "Round", "5": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"decimals": 0, "on_transform": false, "on_inverse": true}, "5": {"lag_1": 7, "method": "Median"}}}
Model Number: 44 with model GLS in generation 0 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}}}
Model Number: 45 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "quadratic", "transformations": {"0": "PctChangeTransformer", "1": "MeanPercentSplitter", "2": "StandardScaler"}, "transformation_params": {"0": {}, "1": {"window": 3}, "2": {}}}
Model Number: 46 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 47 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
Model Number: 48 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 5040, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 0.001, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}}}
Model Number: 49 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 180, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "RobustScaler", "1": "Discretize", "2": "CenterSplit"}, "transformation_params": {"0": {}, "1": {"discretization": "center", "n_bins": 5}, "2": {"fillna": "mean", "center": "zero"}}}
Model Number: 50 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["day", 168, 52], "changepoint_spacing": 90, "changepoint_distance_end": 90, "regression_type": "User", "lambda_": null, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.5, "first_value_only": false, "threshold": 10, "threshold_method": "max"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3605, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 50 in generation 0: BasicLinearModel
Model Number: 51 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "minkowski", "include_differenced": true, "k": 100, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue", "4": "AlignLastValue", "5": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"model": "Linear", "phi": 0.999, "window": 90, "transform_dict": {"fillna": "rolling_mean_24", "transformations": {"0": "RollingMean100thN", "1": "DifferencedTransformer"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "bfill"}}}}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "5": {"rows": 2}}}
Model Number: 52 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "cityblock", "include_differenced": true, "k": 15, "stride_size": 1, "regression_type": "User", "comparison_transformation": {"fillna": "fake_date", "transformations": {"0": "Detrend", "1": "DifferencedTransformer"}, "transformation_params": {"0": {"model": "GLS", "phi": 0.999, "window": null, "transform_dict": null}, "1": {"lag": 1, "fill": "bfill"}}}, "combination_transformation": null} and transformations {"fillna": "time", "transformations": {"0": "ClipOutliers", "1": "Detrend"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": {"fillna": null, "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4}}}}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 1847, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 52 in generation 0: SectionalMotif
Model Number: 53 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "expanded_binarized", "changepoint_spacing": 180, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 0.1, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "DatepartRegression", "1": "AlignLastValue", "2": "SinTrend", "3": "MinMaxScaler", "4": "AlignLastValue"}, "transformation_params": {"0": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 5000}}, "datepart_method": [7, 365.25], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "1": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3627, in fit
np.linalg.inv(X_values.T @ X_values + self.lambda_ * I)
ValueError: matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 1141 is different from 2406)
in model 53 in generation 0: BasicLinearModel
Model Number: 54 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": null, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "STLFilter", "1": "MinMaxScaler", "2": "StandardScaler"}, "transformation_params": {"0": {"decomp_type": "STL", "part": "trend", "seasonal": 7}, "1": {}, "2": {}}}
Model Number: 55 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "pchip", "transformations": {"0": "AlignLastValue", "1": "ThetaTransformer", "2": "DifferencedTransformer", "3": "StandardScaler", "4": "DifferencedTransformer"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "max"}, "1": {"theta_values": [0.8, 1.2]}, "2": {"lag": 1, "fill": "bfill"}, "3": {}, "4": {"lag": 1, "fill": "zero"}}}
Model Number: 56 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "pchip", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "DifferencedTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 365, "transform_dict": null}, "2": {"lag": 1, "fill": "bfill"}, "3": {"rows": 168, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 57 with model AverageValueNaive in generation 0 of 1 with params {"method": "Median", "window": null} and transformations {"fillna": "ffill", "transformations": {"0": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": "quarter"}}}
Model Number: 58 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["morlet_365.25_12_12", "ricker_7_7_1"], "changepoint_spacing": 60, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Detrend", "1": "IntermittentOccurrence", "2": "MeanPercentSplitter"}, "transformation_params": {"0": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "1": {"center": "mean"}, "2": {"window": "forecast_length"}}}
Model Number: 59 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "midhinge", "distance_metric": "correlation", "include_differenced": true, "k": 1, "stride_size": 1, "regression_type": "User", "comparison_transformation": {"fillna": "quadratic", "transformations": {"0": "Log"}, "transformation_params": {"0": {}}}, "combination_transformation": null} and transformations {"fillna": "time", "transformations": {"0": "FFTFilter"}, "transformation_params": {"0": {"cutoff": 0.6, "reverse": false, "on_transform": true, "on_inverse": false}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 1847, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 59 in generation 0: SectionalMotif
Model Number: 60 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 5040, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "quadratic", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 61 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "zero", "transformations": {"0": "AnomalyRemoval", "1": "Constraint", "2": "MinMaxScaler"}, "transformation_params": {"0": {"method": "minmax", "method_params": {"alpha": 0.03}, "fillna": "ffill", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HistoricValues", "1": "PowerTransformer"}, "transformation_params": {"0": {"window": null}, "1": {}}}, "isolated_only": false, "on_inverse": false}, "1": {"constraint_method": "stdev_min", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 3.0, "bounds_only": false, "fillna": "ffill"}, "2": {}}}
Model Number: 62 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 180, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 100, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "pchip", "transformations": {"0": "STLFilter", "1": "AlignLastValue"}, "transformation_params": {"0": {"decomp_type": "seasonal_decompose", "part": "trend"}, "1": {"rows": 84, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 63 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 60, "changepoint_distance_end": 520, "regression_type": null, "lambda_": 1, "trend_phi": 0.97, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}}}
New Generation: 1 of 1
Model Number: 64 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "LevelShiftTransformer"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}}}
Model Number: 65 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "MaxAbsScaler", "1": "QuantileTransformer", "2": "HistoricValues", "3": "ThetaTransformer", "4": "ReplaceConstant"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"window": null}, "3": {"theta_values": [0.4, 1.6]}, "4": {"constant": 0, "reintroduction_model": {"model": "KNN", "model_params": {"n_neighbors": 5, "weights": "uniform", "p": 2, "leaf_size": 30}, "datepart_method": ["common_fourier"]}, "fillna": null}}}
Model Number: 66 with model SeasonalityMotif in generation 1 of 1 with params {"window": 50, "point_method": "trimmed_mean_40", "distance_metric": "mqae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "ThetaTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"theta_values": [0.6, 1.4]}}}
Model Number: 67 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "canberra", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}, "1": {"rows": 1, "lag": 168, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}}}
Model Number: 68 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "akima", "transformations": {"0": "RollingMeanTransformer", "1": "QuantileTransformer", "2": "Discretize", "3": "Constraint"}, "transformation_params": {"0": {"fixed": true, "window": 28, "macro_micro": false, "center": true}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"discretization": "sklearn-quantile", "n_bins": 50}, "3": {"constraint_method": "dampening", "constraint_direction": "upper", "constraint_regularization": 1.0, "constraint_value": 0.99, "bounds_only": false, "fillna": null}}}
Model Number: 69 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 60, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": 0.97, "holiday_countries_used": false} and transformations {"fillna": "mean", "transformations": {"0": "SeasonalDifference", "1": "QuantileTransformer"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 70 with model SectionalMotif in generation 1 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "braycurtis", "include_differenced": true, "k": 20, "stride_size": 5, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 71 with model ConstantNaive in generation 1 of 1 with params {"constant": -1} and transformations {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer", "2": "AnomalyRemoval"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"method": "rolling_zscore", "method_params": {"distribution": "norm", "alpha": 0.05, "rolling_periods": 200, "center": true}, "fillna": "ffill", "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}, "isolated_only": false, "on_inverse": false}}}
Model Number: 72 with model SectionalMotif in generation 1 of 1 with params {"window": 10, "point_method": "mean", "distance_metric": "sqeuclidean", "include_differenced": true, "k": 20, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "EWMAFilter"}, "transformation_params": {"0": {"rows": null, "displacement_rows": 4, "quantile": 1.0, "decay_span": null}, "1": {"span": 12}}}, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PositiveShift", "1": "QuantileTransformer", "2": "RobustScaler", "3": "EWMAFilter", "4": "RollingMeanTransformer", "5": "DatepartRegression"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}, "3": {"span": 7}, "4": {"fixed": true, "window": 7, "macro_micro": false, "center": true}, "5": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 2}}, "datepart_method": "expanded", "polynomial_degree": null, "transform_dict": {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"rows": 7, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}, "holiday_countries_used": false, "lags": 1, "forward_lags": null}}}
Model Number: 73 with model GLS in generation 1 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "median", "transformations": {"0": "SeasonalDifference", "1": "QuantileTransformer", "2": "SeasonalDifference", "3": "RobustScaler", "4": "AlignLastValue"}, "transformation_params": {"0": {"lag_1": 12, "method": "Mean"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 65, "method": "Median"}, "3": {}, "4": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}
Model Number: 74 with model SeasonalityMotif in generation 1 of 1 with params {"window": 30, "point_method": "weighted_mean", "distance_metric": "mae", "k": 10, "datepart_method": [7, 365.25], "independent": false} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"method": "remove", "std_threshold": 4, "fillna": "ffill"}}}
Model Number: 75 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "canberra", "k": 10, "datepart_method": "common_fourier", "independent": true} and transformations {"fillna": "quadratic", "transformations": {"0": "Detrend", "1": "DatepartRegression"}, "transformation_params": {"0": {"model": "Linear", "phi": 1, "window": 365, "transform_dict": null}, "1": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": ["weekdaymonthofyear", "quarter", "dayofweek"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}}
Model Number: 76 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "mean", "transformations": {"0": "DifferencedTransformer"}, "transformation_params": {"0": {"lag": 1, "fill": "zero"}}}
Model Number: 77 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 78 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "MinMaxScaler", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 365, "transform_dict": null}, "2": {}, "3": {"rows": 168, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 79 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "HolidayTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"threshold": 0.8, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": false, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": true, "use_hindu_holidays": false, "anomaly_detector_params": {"method": "med_diff", "method_params": {"distribution": "beta", "alpha": 0.05}, "fillna": "ffill", "transform_dict": null, "isolated_only": true, "on_inverse": false}, "remove_excess_anomalies": false, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": [168, "hour"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {"rows": 1, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 80 with model AverageValueNaive in generation 1 of 1 with params {"method": "Weighted_Mean", "window": null} and transformations {"fillna": "ffill", "transformations": {"0": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 285}}}
Model Number: 81 with model GLS in generation 1 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": 180, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "ChangepointDetrend", "1": "PctChangeTransformer", "2": "BKBandpassFilter", "3": "DifferencedTransformer"}, "transformation_params": {"0": {"model": "Poisson", "changepoint_spacing": 5040, "changepoint_distance_end": 28, "datepart_method": "simple"}, "1": {}, "2": {"low": 7, "high": 90, "K": 6, "lanczos_factor": false, "return_diff": true, "on_transform": true, "on_inverse": false}, "3": {"lag": 7, "fill": "bfill"}}}
Model Number: 82 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "midhinge", "distance_metric": "hamming", "include_differenced": true, "k": 3, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "fake_date", "transformations": {"0": "FFTDecomposition"}, "transformation_params": {"0": {"n_harmonics": 0.5, "detrend": null}}}, "combination_transformation": null} and transformations {"fillna": "fake_date", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/linear_model/_linear_loss.py:330: RuntimeWarning: overflow encountered in matmul
grad[:n_features] = X.T @ grad_pointwise + l2_reg_strength * weights
/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/linear_model/_linear_loss.py:330: RuntimeWarning: invalid value encountered in matmul
grad[:n_features] = X.T @ grad_pointwise + l2_reg_strength * weights
/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/linear_model/_linear_loss.py:330: RuntimeWarning: overflow encountered in matmul
grad[:n_features] = X.T @ grad_pointwise + l2_reg_strength * weights
/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/linear_model/_linear_loss.py:330: RuntimeWarning: invalid value encountered in matmul
grad[:n_features] = X.T @ grad_pointwise + l2_reg_strength * weights
Model Number: 83 with model SectionalMotif in generation 1 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "cosine", "include_differenced": true, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "AlignLastValue", "1": "ScipyFilter", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "1": {"method": "savgol_filter", "method_args": {"window_length": 7, "polyorder": 2, "deriv": 0, "mode": "interp"}}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 84 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "zero", "transformations": {"0": "ClipOutliers", "1": "Constraint", "2": "MinMaxScaler"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"constraint_method": "stdev_min", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 3.0, "bounds_only": false, "fillna": "ffill"}, "2": {}}}
Model Number: 85 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "zero", "transformations": {"0": "AlignLastValue", "1": "Constraint", "2": "FFTFilter"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "1": {"constraint_method": "stdev_min", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 3.0, "bounds_only": false, "fillna": "ffill"}, "2": {"cutoff": 0.2, "reverse": false, "on_transform": false, "on_inverse": false}}}
Model Number: 86 with model SectionalMotif in generation 1 of 1 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": true, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 87 with model SectionalMotif in generation 1 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "sqeuclidean", "include_differenced": true, "k": 100, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "Slice"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"method": 100}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1598, in model_forecast
return model.predict(
^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 943, in predict
df_forecast = self.model.predict(
^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 2002, in predict
res_idx = np.argpartition(res_sum, num_top, axis=0)[0:num_top]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/numpy/_core/fromnumeric.py", line 908, in argpartition
return _wrapfunc(a, 'argpartition', kth, axis=axis, kind=kind, order=order)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/numpy/_core/fromnumeric.py", line 57, in _wrapfunc
return bound(*args, **kwds)
^^^^^^^^^^^^^^^^^^^^
ValueError: kth(=100) out of bounds (66)
in model 87 in generation 1: SectionalMotif
Model Number: 88 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 520, "regression_type": "User", "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "pchip", "transformations": {"0": "STLFilter", "1": "AlignLastValue", "2": "RollingMean100thN", "3": "RobustScaler", "4": "StandardScaler", "5": "SeasonalDifference"}, "transformation_params": {"0": {"decomp_type": "seasonal_decompose", "part": "trend"}, "1": {"rows": 84, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {"lag_1": 24, "method": "LastValue"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3605, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 88 in generation 1: BasicLinearModel
Model Number: 89 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "mean", "transformations": {"0": "PctChangeTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 90 with model GLS in generation 1 of 1 with params {"changepoint_spacing": 180, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "akima", "transformations": {"0": "ClipOutliers", "1": "Discretize", "2": "Log"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"discretization": "lower", "n_bins": 50}, "2": {}}}
Model Number: 91 with model GLS in generation 1 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": 90, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "Discretize", "3": "AlignLastValue", "4": "AlignLastValue", "5": "RollingMeanTransformer"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"discretization": "upper", "n_bins": 10}, "3": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "5": {"fixed": true, "window": 3, "macro_micro": false, "center": true}}}
Model Number: 92 with model SeasonalityMotif in generation 1 of 1 with params {"window": 15, "point_method": "trimmed_mean_20", "distance_metric": "mqae", "k": 10, "datepart_method": "simple_2", "independent": false} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
Model Number: 93 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "SinTrend"}, "transformation_params": {"0": {"lag": 1, "fill": "bfill"}, "1": {}}}
Model Number: 94 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 100, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "CenterSplit", "1": "LocalLinearTrend", "2": "AlignLastValue", "3": "LevelShiftTransformer", "4": "AlignLastValue", "5": "MaxAbsScaler"}, "transformation_params": {"0": {"fillna": "akima", "center": "zero"}, "1": {"rolling_window": 0.1, "n_tails": 0.1, "n_future": 0.2, "method": "mean", "macro_micro": true}, "2": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "5": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6696, in inverse_transform
df = self._inverse_one(
^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6666, in _inverse_one
df = self.transformers[i].inverse_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 3594, in inverse_transform
micro = micro.rename(columns=lambda x: str(x)[: -len(self.suffix)])[
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/pandas/core/frame.py", line 4119, in __getitem__
indexer = self.columns._get_indexer_strict(key, "columns")[1]
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/pandas/core/indexes/base.py", line 6212, in _get_indexer_strict
self._raise_if_missing(keyarr, indexer, axis_name)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/pandas/core/indexes/base.py", line 6264, in _raise_if_missing
raise KeyError(f"{not_found} not in index")
KeyError: "['theta', 'kappa', 'rho', 'sigma'] not in index"
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1598, in model_forecast
return model.predict(
^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 960, in predict
df_forecast.forecast = self.transformer_object.inverse_transform(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6703, in inverse_transform
raise Exception(err_str) from e
Exception: Transformer LocalLinearTrend failed on inverse from params ffill_mean_biased {'0': {'fillna': 'akima', 'center': 'zero'}, '1': {'rolling_window': 0.1, 'n_tails': 0.1, 'n_future': 0.2, 'method': 'mean', 'macro_micro': True}, '2': {'rows': 1, 'lag': 1, 'method': 'multiplicative', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '3': {'window_size': 30, 'alpha': 2.0, 'grouping_forward_limit': 4, 'max_level_shifts': 10, 'alignment': 'average'}, '4': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': None, 'threshold_method': 'mean'}, '5': {}} with KeyError("['theta', 'kappa', 'rho', 'sigma'] not in index")
in model 94 in generation 1: SectionalMotif
Model Number: 95 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "DifferencedTransformer"}, "transformation_params": {"0": {"lag": 7, "fill": "zero"}}}
Model Number: 96 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 7, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "SeasonalDifference", "1": "ClipOutliers"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 97 with model AverageValueNaive in generation 1 of 1 with params {"method": "Median", "window": 7} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "SinTrend"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {}}}
Model Number: 98 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "rolling_mean", "transformations": {"0": "AnomalyRemoval", "1": "Constraint", "2": "MinMaxScaler"}, "transformation_params": {"0": {"method": "minmax", "method_params": {"alpha": 0.03}, "fillna": "ffill", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HistoricValues", "1": "PowerTransformer"}, "transformation_params": {"0": {"window": null}, "1": {}}}, "isolated_only": false, "on_inverse": false}, "1": {"constraint_method": "stdev_min", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 3.0, "bounds_only": false, "fillna": "ffill"}, "2": {}}}
Model Number: 99 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "linear", "transformations": {"0": "EWMAFilter", "1": "AlignLastValue", "2": "AlignLastValue", "3": "Detrend", "4": "MaxAbsScaler", "5": "AlignLastValue"}, "transformation_params": {"0": {"span": 7}, "1": {"rows": 7, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "3": {"model": "GLS", "phi": 1, "window": 900, "transform_dict": null}, "4": {}, "5": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}
Model Number: 100 with model AverageValueNaive in generation 1 of 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "pchip", "transformations": {"0": "ClipOutliers", "1": "ShiftFirstValue", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"rows": 1}, "2": {"lag_1": 364, "method": "Mean"}, "3": {"window_size": 7, "alpha": 2.5, "grouping_forward_limit": 2, "max_level_shifts": 10, "alignment": "rolling_diff_3nn"}}}
Model Number: 101 with model SeasonalNaive in generation 1 of 1 with params {"method": "median", "lag_1": 96, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "Round"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 102 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "mean", "transformations": {"0": "RegressionFilter"}, "transformation_params": {"0": {"sigma": 1, "rolling_window": 90, "run_order": "season_first", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 2000}}, "datepart_method": "simple_2", "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "holiday_params": null, "trend_method": "local_linear"}}}
Model Number: 103 with model GLS in generation 1 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
Validation Round: 1
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-02-28', '2023-03-01', '2023-03-02', '2023-03-03',
'2023-03-04', '2023-03-05', '2023-03-06', '2023-03-07',
'2023-03-08', '2023-03-09'],
dtype='datetime64[ns]', name='Date', length=1111, freq=None)
Model Number: 1 of 19 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
📈 1 - LastValueNaive with avg smape 35.43:
Model Number: 2 of 19 with model AverageValueNaive for Validation 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 0.05}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
2 - AverageValueNaive with avg smape 36.13:
Model Number: 3 of 19 with model SeasonalityMotif for Validation 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "mae", "k": 10, "datepart_method": "common_fourier", "independent": false} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
3 - SeasonalityMotif with avg smape 36.93:
Model Number: 4 of 19 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
4 - LastValueNaive with avg smape 37.15:
Model Number: 5 of 19 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
5 - ConstantNaive with avg smape 36.13:
Model Number: 6 of 19 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
6 - LastValueNaive with avg smape 36.13:
Model Number: 7 of 19 with model SeasonalityMotif for Validation 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "mean", "transformations": {"0": "DifferencedTransformer"}, "transformation_params": {"0": {"lag": 1, "fill": "zero"}}}
7 - SeasonalityMotif with avg smape 36.13:
Model Number: 8 of 19 with model AverageValueNaive for Validation 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
8 - AverageValueNaive with avg smape 40.0:
Model Number: 9 of 19 with model SectionalMotif for Validation 1 with params {"window": 7, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
9 - SectionalMotif with avg smape 50.24:
Model Number: 10 of 19 with model SectionalMotif for Validation 1 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": true, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
10 - SectionalMotif with avg smape 63.25:
Model Number: 11 of 19 with model AverageValueNaive for Validation 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "SinTrend"}, "transformation_params": {"0": {}, "1": {}}}
11 - AverageValueNaive with avg smape 72.61:
Model Number: 12 of 19 with model SectionalMotif for Validation 1 with params {"window": 10, "point_method": "mean", "distance_metric": "sqeuclidean", "include_differenced": true, "k": 20, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "EWMAFilter"}, "transformation_params": {"0": {"rows": null, "displacement_rows": 4, "quantile": 1.0, "decay_span": null}, "1": {"span": 12}}}, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PositiveShift", "1": "QuantileTransformer", "2": "RobustScaler", "3": "EWMAFilter", "4": "RollingMeanTransformer", "5": "DatepartRegression"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}, "3": {"span": 7}, "4": {"fixed": true, "window": 7, "macro_micro": false, "center": true}, "5": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 2}}, "datepart_method": "expanded", "polynomial_degree": null, "transform_dict": {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"rows": 7, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}, "holiday_countries_used": false, "lags": 1, "forward_lags": null}}}
12 - SectionalMotif with avg smape 68.81:
Model Number: 13 of 19 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "quadratic", "transformations": {"0": "PctChangeTransformer", "1": "MeanPercentSplitter", "2": "StandardScaler"}, "transformation_params": {"0": {}, "1": {"window": 3}, "2": {}}}
13 - ConstantNaive with avg smape 47.81:
Model Number: 14 of 19 with model SeasonalityMotif for Validation 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
14 - SeasonalityMotif with avg smape 35.69:
Model Number: 15 of 19 with model GLS for Validation 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
📈 15 - GLS with avg smape 33.18:
Model Number: 16 of 19 with model GLS for Validation 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
16 - GLS with avg smape 33.18:
Model Number: 17 of 19 with model BasicLinearModel for Validation 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "LevelShiftTransformer"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}}}
17 - BasicLinearModel with avg smape 79.91:
Model Number: 18 of 19 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "HolidayTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"threshold": 0.8, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": false, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": true, "use_hindu_holidays": false, "anomaly_detector_params": {"method": "med_diff", "method_params": {"distribution": "beta", "alpha": 0.05}, "fillna": "ffill", "transform_dict": null, "isolated_only": true, "on_inverse": false}, "remove_excess_anomalies": false, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": [168, "hour"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {"rows": 1, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
18 - LastValueNaive with avg smape 76.73:
Model Number: 19 of 19 with model SeasonalityMotif for Validation 1 with params {"window": 15, "point_method": "trimmed_mean_20", "distance_metric": "mqae", "k": 10, "datepart_method": "simple_2", "independent": false} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
19 - SeasonalityMotif with avg smape 53.54:
Validation Round: 2
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-01-29', '2023-01-30', '2023-01-31', '2023-02-01',
'2023-02-02', '2023-02-03', '2023-02-04', '2023-02-05',
'2023-02-06', '2023-02-07'],
dtype='datetime64[ns]', name='Date', length=1081, freq=None)
Model Number: 1 of 19 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
📈 1 - LastValueNaive with avg smape 52.6:
Model Number: 2 of 19 with model AverageValueNaive for Validation 2 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 0.05}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
📈 2 - AverageValueNaive with avg smape 44.95:
Model Number: 3 of 19 with model SeasonalityMotif for Validation 2 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "mae", "k": 10, "datepart_method": "common_fourier", "independent": false} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
3 - SeasonalityMotif with avg smape 65.52:
Model Number: 4 of 19 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
4 - LastValueNaive with avg smape 52.13:
Model Number: 5 of 19 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
5 - ConstantNaive with avg smape 52.86:
Model Number: 6 of 19 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
6 - LastValueNaive with avg smape 52.86:
Model Number: 7 of 19 with model SeasonalityMotif for Validation 2 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "mean", "transformations": {"0": "DifferencedTransformer"}, "transformation_params": {"0": {"lag": 1, "fill": "zero"}}}
7 - SeasonalityMotif with avg smape 52.86:
Model Number: 8 of 19 with model AverageValueNaive for Validation 2 with params {"method": "Mean", "window": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
8 - AverageValueNaive with avg smape 54.75:
Model Number: 9 of 19 with model SectionalMotif for Validation 2 with params {"window": 7, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
9 - SectionalMotif with avg smape 62.6:
Model Number: 10 of 19 with model SectionalMotif for Validation 2 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": true, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
10 - SectionalMotif with avg smape 59.49:
Model Number: 11 of 19 with model AverageValueNaive for Validation 2 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "SinTrend"}, "transformation_params": {"0": {}, "1": {}}}
11 - AverageValueNaive with avg smape 70.35:
Model Number: 12 of 19 with model SectionalMotif for Validation 2 with params {"window": 10, "point_method": "mean", "distance_metric": "sqeuclidean", "include_differenced": true, "k": 20, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "EWMAFilter"}, "transformation_params": {"0": {"rows": null, "displacement_rows": 4, "quantile": 1.0, "decay_span": null}, "1": {"span": 12}}}, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PositiveShift", "1": "QuantileTransformer", "2": "RobustScaler", "3": "EWMAFilter", "4": "RollingMeanTransformer", "5": "DatepartRegression"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}, "3": {"span": 7}, "4": {"fixed": true, "window": 7, "macro_micro": false, "center": true}, "5": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 2}}, "datepart_method": "expanded", "polynomial_degree": null, "transform_dict": {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"rows": 7, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}, "holiday_countries_used": false, "lags": 1, "forward_lags": null}}}
12 - SectionalMotif with avg smape 68.08:
Model Number: 13 of 19 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "quadratic", "transformations": {"0": "PctChangeTransformer", "1": "MeanPercentSplitter", "2": "StandardScaler"}, "transformation_params": {"0": {}, "1": {"window": 3}, "2": {}}}
📈 13 - ConstantNaive with avg smape 36.91:
Model Number: 14 of 19 with model SeasonalityMotif for Validation 2 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
14 - SeasonalityMotif with avg smape 52.88:
Model Number: 15 of 19 with model GLS for Validation 2 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
📈 15 - GLS with avg smape 32.81:
Model Number: 16 of 19 with model GLS for Validation 2 with params {"changepoint_spacing": 60, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
16 - GLS with avg smape 32.81:
Model Number: 17 of 19 with model BasicLinearModel for Validation 2 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "LevelShiftTransformer"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}}}
17 - BasicLinearModel with avg smape 100.1:
Model Number: 18 of 19 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "HolidayTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"threshold": 0.8, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": false, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": true, "use_hindu_holidays": false, "anomaly_detector_params": {"method": "med_diff", "method_params": {"distribution": "beta", "alpha": 0.05}, "fillna": "ffill", "transform_dict": null, "isolated_only": true, "on_inverse": false}, "remove_excess_anomalies": false, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": [168, "hour"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {"rows": 1, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
18 - LastValueNaive with avg smape 93.91:
Model Number: 19 of 19 with model SeasonalityMotif for Validation 2 with params {"window": 15, "point_method": "trimmed_mean_20", "distance_metric": "mqae", "k": 10, "datepart_method": "simple_2", "independent": false} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
19 - SeasonalityMotif with avg smape 76.41:
Validation Round: 0
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-03-27', '2023-03-28', '2023-03-29', '2023-03-30',
'2023-03-31', '2023-04-01', '2023-04-02', '2023-04-03',
'2023-04-04', '2023-04-05'],
dtype='datetime64[ns]', name='Date', length=1138, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 1 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 4, "model_metric": "Score", "models": {"70f35596ceb2e895d60d3da645b45a56": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 5, \"point_method\": \"weighted_mean\", \"distance_metric\": \"mae\", \"k\": 10, \"datepart_method\": \"common_fourier\", \"independent\": false}", "TransformationParameters": "{\"fillna\": \"nearest\", \"transformations\": {\"0\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": false}}}"}, "1b65dbe1ca2a4164c7c6daeeeb897e2b": {"Model": "AverageValueNaive", "ModelParameters": "{\"method\": \"trimmed_mean_40\", \"window\": 28}", "TransformationParameters": "{\"fillna\": \"fake_date\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"PositiveShift\", \"2\": \"HolidayTransformer\", \"3\": \"PositiveShift\", \"4\": \"StandardScaler\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"max\"}, \"1\": {}, \"2\": {\"threshold\": 0.9, \"splash_threshold\": null, \"use_dayofmonth_holidays\": true, \"use_wkdom_holidays\": true, \"use_wkdeom_holidays\": false, \"use_lunar_holidays\": false, \"use_lunar_weekday\": false, \"use_islamic_holidays\": false, \"use_hebrew_holidays\": false, \"use_hindu_holidays\": true, \"anomaly_detector_params\": {\"method\": \"IQR\", \"method_params\": {\"iqr_threshold\": 2.0, \"iqr_quantiles\": [0.25, 0.75]}, \"fillna\": \"fake_date\", \"transform_dict\": {\"fillna\": \"ffill\", \"transformations\": {\"0\": \"HPFilter\", \"1\": \"ClipOutliers\"}, \"transformation_params\": {\"0\": {\"part\": \"trend\", \"lamb\": 129600}, \"1\": {\"method\": \"clip\", \"std_threshold\": 3, \"fillna\": null}}}, \"isolated_only\": false, \"on_inverse\": false}, \"remove_excess_anomalies\": true, \"impact\": \"datepart_regression\", \"regression_params\": {\"regression_model\": {\"model\": \"DecisionTree\", \"model_params\": {\"max_depth\": 3, \"min_samples_split\": 0.05}}, \"datepart_method\": \"recurring\", \"polynomial_degree\": 2, \"transform_dict\": {\"fillna\": null, \"transformations\": {\"0\": \"bkfilter\"}, \"transformation_params\": {\"0\": {}}}, \"holiday_countries_used\": false, \"lags\": null, \"forward_lags\": null}}, \"3\": {}, \"4\": {}}}"}, "7511845b30dd1d33b9cf7cf5b0b914ca": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"akima\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"AlignLastValue\", \"2\": \"bkfilter\", \"3\": \"PositiveShift\", \"4\": \"StandardScaler\", \"5\": \"bkfilter\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 28, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": true, \"threshold\": 10, \"threshold_method\": \"mean\"}, \"1\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 10, \"threshold_method\": \"mean\"}, \"2\": {}, \"3\": {}, \"4\": {}, \"5\": {}}}"}, "01b50f79d962886b0e050af50362d950": {"Model": "SectionalMotif", "ModelParameters": "{\"window\": 7, \"point_method\": \"midhinge\", \"distance_metric\": \"canberra\", \"include_differenced\": true, \"k\": 5, \"stride_size\": 1, \"regression_type\": null, \"comparison_transformation\": null, \"combination_transformation\": null}", "TransformationParameters": "{\"fillna\": \"mean\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"QuantileTransformer\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 1, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 1000}}}"}}, "series": {"theta": "7511845b30dd1d33b9cf7cf5b0b914ca", "kappa": "1b65dbe1ca2a4164c7c6daeeeb897e2b", "rho": "01b50f79d962886b0e050af50362d950", "sigma": "70f35596ceb2e895d60d3da645b45a56"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 4 SeasonalityMotif started
Ensemble Horizontal component 2 of 4 AverageValueNaive started
Ensemble Horizontal component 3 of 4 LastValueNaive started
Ensemble Horizontal component 4 of 4 SectionalMotif started
📈 1 - Ensemble with avg smape 57.32:
📈 2 - Ensemble with avg smape 56.65:
📈 3 - Ensemble with avg smape 53.34:
📈 4 - Ensemble with avg smape 53.19:
5 - Ensemble with avg smape 55.67:
6 - Ensemble with avg smape 55.7:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Validation Round: 1
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-02-26', '2023-02-27', '2023-02-28', '2023-03-01',
'2023-03-02', '2023-03-03', '2023-03-04', '2023-03-05',
'2023-03-06', '2023-03-07'],
dtype='datetime64[ns]', name='Date', length=1109, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 2 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 4, "model_metric": "Score", "models": {"70f35596ceb2e895d60d3da645b45a56": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 5, \"point_method\": \"weighted_mean\", \"distance_metric\": \"mae\", \"k\": 10, \"datepart_method\": \"common_fourier\", \"independent\": false}", "TransformationParameters": "{\"fillna\": \"nearest\", \"transformations\": {\"0\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": false}}}"}, "1b65dbe1ca2a4164c7c6daeeeb897e2b": {"Model": "AverageValueNaive", "ModelParameters": "{\"method\": \"trimmed_mean_40\", \"window\": 28}", "TransformationParameters": "{\"fillna\": \"fake_date\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"PositiveShift\", \"2\": \"HolidayTransformer\", \"3\": \"PositiveShift\", \"4\": \"StandardScaler\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"max\"}, \"1\": {}, \"2\": {\"threshold\": 0.9, \"splash_threshold\": null, \"use_dayofmonth_holidays\": true, \"use_wkdom_holidays\": true, \"use_wkdeom_holidays\": false, \"use_lunar_holidays\": false, \"use_lunar_weekday\": false, \"use_islamic_holidays\": false, \"use_hebrew_holidays\": false, \"use_hindu_holidays\": true, \"anomaly_detector_params\": {\"method\": \"IQR\", \"method_params\": {\"iqr_threshold\": 2.0, \"iqr_quantiles\": [0.25, 0.75]}, \"fillna\": \"fake_date\", \"transform_dict\": {\"fillna\": \"ffill\", \"transformations\": {\"0\": \"HPFilter\", \"1\": \"ClipOutliers\"}, \"transformation_params\": {\"0\": {\"part\": \"trend\", \"lamb\": 129600}, \"1\": {\"method\": \"clip\", \"std_threshold\": 3, \"fillna\": null}}}, \"isolated_only\": false, \"on_inverse\": false}, \"remove_excess_anomalies\": true, \"impact\": \"datepart_regression\", \"regression_params\": {\"regression_model\": {\"model\": \"DecisionTree\", \"model_params\": {\"max_depth\": 3, \"min_samples_split\": 0.05}}, \"datepart_method\": \"recurring\", \"polynomial_degree\": 2, \"transform_dict\": {\"fillna\": null, \"transformations\": {\"0\": \"bkfilter\"}, \"transformation_params\": {\"0\": {}}}, \"holiday_countries_used\": false, \"lags\": null, \"forward_lags\": null}}, \"3\": {}, \"4\": {}}}"}, "7511845b30dd1d33b9cf7cf5b0b914ca": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"akima\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"AlignLastValue\", \"2\": \"bkfilter\", \"3\": \"PositiveShift\", \"4\": \"StandardScaler\", \"5\": \"bkfilter\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 28, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": true, \"threshold\": 10, \"threshold_method\": \"mean\"}, \"1\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 10, \"threshold_method\": \"mean\"}, \"2\": {}, \"3\": {}, \"4\": {}, \"5\": {}}}"}, "01b50f79d962886b0e050af50362d950": {"Model": "SectionalMotif", "ModelParameters": "{\"window\": 7, \"point_method\": \"midhinge\", \"distance_metric\": \"canberra\", \"include_differenced\": true, \"k\": 5, \"stride_size\": 1, \"regression_type\": null, \"comparison_transformation\": null, \"combination_transformation\": null}", "TransformationParameters": "{\"fillna\": \"mean\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"QuantileTransformer\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 1, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 1000}}}"}}, "series": {"theta": "7511845b30dd1d33b9cf7cf5b0b914ca", "kappa": "1b65dbe1ca2a4164c7c6daeeeb897e2b", "rho": "01b50f79d962886b0e050af50362d950", "sigma": "70f35596ceb2e895d60d3da645b45a56"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 4 SeasonalityMotif started
Ensemble Horizontal component 2 of 4 AverageValueNaive started
Ensemble Horizontal component 3 of 4 LastValueNaive started
Ensemble Horizontal component 4 of 4 SectionalMotif started
📈 1 - Ensemble with avg smape 30.76:
📈 2 - Ensemble with avg smape 30.47:
3 - Ensemble with avg smape 44.04:
📈 4 - Ensemble with avg smape 30.28:
5 - Ensemble with avg smape 32.0:
6 - Ensemble with avg smape 40.84:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Validation Round: 2
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-01-28', '2023-01-29', '2023-01-30', '2023-01-31',
'2023-02-01', '2023-02-02', '2023-02-03', '2023-02-04',
'2023-02-05', '2023-02-06'],
dtype='datetime64[ns]', name='Date', length=1080, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 3 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 4, "model_metric": "Score", "models": {"70f35596ceb2e895d60d3da645b45a56": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 5, \"point_method\": \"weighted_mean\", \"distance_metric\": \"mae\", \"k\": 10, \"datepart_method\": \"common_fourier\", \"independent\": false}", "TransformationParameters": "{\"fillna\": \"nearest\", \"transformations\": {\"0\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": false}}}"}, "1b65dbe1ca2a4164c7c6daeeeb897e2b": {"Model": "AverageValueNaive", "ModelParameters": "{\"method\": \"trimmed_mean_40\", \"window\": 28}", "TransformationParameters": "{\"fillna\": \"fake_date\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"PositiveShift\", \"2\": \"HolidayTransformer\", \"3\": \"PositiveShift\", \"4\": \"StandardScaler\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"max\"}, \"1\": {}, \"2\": {\"threshold\": 0.9, \"splash_threshold\": null, \"use_dayofmonth_holidays\": true, \"use_wkdom_holidays\": true, \"use_wkdeom_holidays\": false, \"use_lunar_holidays\": false, \"use_lunar_weekday\": false, \"use_islamic_holidays\": false, \"use_hebrew_holidays\": false, \"use_hindu_holidays\": true, \"anomaly_detector_params\": {\"method\": \"IQR\", \"method_params\": {\"iqr_threshold\": 2.0, \"iqr_quantiles\": [0.25, 0.75]}, \"fillna\": \"fake_date\", \"transform_dict\": {\"fillna\": \"ffill\", \"transformations\": {\"0\": \"HPFilter\", \"1\": \"ClipOutliers\"}, \"transformation_params\": {\"0\": {\"part\": \"trend\", \"lamb\": 129600}, \"1\": {\"method\": \"clip\", \"std_threshold\": 3, \"fillna\": null}}}, \"isolated_only\": false, \"on_inverse\": false}, \"remove_excess_anomalies\": true, \"impact\": \"datepart_regression\", \"regression_params\": {\"regression_model\": {\"model\": \"DecisionTree\", \"model_params\": {\"max_depth\": 3, \"min_samples_split\": 0.05}}, \"datepart_method\": \"recurring\", \"polynomial_degree\": 2, \"transform_dict\": {\"fillna\": null, \"transformations\": {\"0\": \"bkfilter\"}, \"transformation_params\": {\"0\": {}}}, \"holiday_countries_used\": false, \"lags\": null, \"forward_lags\": null}}, \"3\": {}, \"4\": {}}}"}, "7511845b30dd1d33b9cf7cf5b0b914ca": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"akima\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"AlignLastValue\", \"2\": \"bkfilter\", \"3\": \"PositiveShift\", \"4\": \"StandardScaler\", \"5\": \"bkfilter\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 28, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": true, \"threshold\": 10, \"threshold_method\": \"mean\"}, \"1\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 10, \"threshold_method\": \"mean\"}, \"2\": {}, \"3\": {}, \"4\": {}, \"5\": {}}}"}, "01b50f79d962886b0e050af50362d950": {"Model": "SectionalMotif", "ModelParameters": "{\"window\": 7, \"point_method\": \"midhinge\", \"distance_metric\": \"canberra\", \"include_differenced\": true, \"k\": 5, \"stride_size\": 1, \"regression_type\": null, \"comparison_transformation\": null, \"combination_transformation\": null}", "TransformationParameters": "{\"fillna\": \"mean\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"QuantileTransformer\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 1, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 1000}}}"}}, "series": {"theta": "7511845b30dd1d33b9cf7cf5b0b914ca", "kappa": "1b65dbe1ca2a4164c7c6daeeeb897e2b", "rho": "01b50f79d962886b0e050af50362d950", "sigma": "70f35596ceb2e895d60d3da645b45a56"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 4 SeasonalityMotif started
Ensemble Horizontal component 2 of 4 AverageValueNaive started
Ensemble Horizontal component 3 of 4 LastValueNaive started
Ensemble Horizontal component 4 of 4 SectionalMotif started
📈 1 - Ensemble with avg smape 58.62:
📈 2 - Ensemble with avg smape 57.39:
3 - Ensemble with avg smape 68.22:
📈 4 - Ensemble with avg smape 55.81:
5 - Ensemble with avg smape 61.63:
📈 6 - Ensemble with avg smape 50.54:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Ensemble Horizontal component 1 of 4 SeasonalityMotif started
Ensemble Horizontal component 2 of 4 AverageValueNaive started
Ensemble Horizontal component 3 of 4 LastValueNaive started
Ensemble Horizontal component 4 of 4 SectionalMotif started
2025-11-22 at 12:42:16 | ERROR | validation of end failed
Check the predicted ODE parameter values.
[6]:
df = snr.append().summary()
df.loc[df["Start"] >= future_start_date]
2025-11-22 at 12:42:16 | ERROR | validation of end failed
2025-11-22 at 12:42:16 | ERROR | validation of end failed
[6]:
| Start | End | Rt | theta | kappa | rho | sigma | alpha1 [-] | 1/alpha2 [day] | 1/beta [day] | 1/gamma [day] | ODE | tau | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Scenario | Phase | |||||||||||||
| Predicted | 68th | 2023-05-09 | 2023-06-07 | 0.95 | 0.000246 | 0.000002 | 0.001034 | 0.00109 | 0.0 | 8752 | 16 | 15 | SIR-F Model | 24 |
Check the dynamics.
[7]:
snr.simulate(name="Predicted");
2. Time-series prediction with indicators
using `future_regressor functionality of AutoTS <https://winedarksea.github.io/AutoTS/build/html/source/tutorial.html#adding-regressors-and-other-information>`__, we will predict ODE parameter values with indicators. We can download/create time-series data of indicators using DataEngineer class as explained in Tutorial: Data preparation and Tutorial: Data
engineering.
[8]:
data_eng = cs.DataEngineer()
data_eng.download(databases=["japan", "covid19dh", "owid"]).clean().transform()
subset_df, *_ = data_eng.subset(geo="Japan")
indicator_df = subset_df.drop(["Population", "Susceptible", "Confirmed", "Infected", "Fatal", "Recovered"], axis=1)
indicator_df
[8]:
| Cancel_events | Contact_tracing | Gatherings_restrictions | Information_campaigns | Internal_movement_restrictions | International_movement_restrictions | School_closing | Stay_home_restrictions | Stringency_index | Testing_policy | ... | Transport_closing | Vaccinated_full | Vaccinated_once | Vaccinations | Vaccinations_boosters | Workplace_closing | Country_0 | Country_Japan | Product_0 | Product_Moderna, Novavax, Oxford/AstraZeneca, Pfizer/BioNTech | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Date | |||||||||||||||||||||
| 2020-02-05 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | ... | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 1 | 0 | 1 | 0 |
| 2020-02-06 | 0.0 | 2.0 | 0.0 | 2.0 | 0.0 | 3.0 | 0.0 | 0.0 | 19.44 | 1.0 | ... | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 1 | 1 | 0 |
| 2020-02-07 | 0.0 | 2.0 | 0.0 | 2.0 | 0.0 | 3.0 | 0.0 | 0.0 | 19.44 | 1.0 | ... | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 1 | 1 | 0 |
| 2020-02-08 | 0.0 | 2.0 | 0.0 | 2.0 | 0.0 | 3.0 | 0.0 | 0.0 | 19.44 | 1.0 | ... | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 1 | 1 | 0 |
| 2020-02-09 | 0.0 | 2.0 | 0.0 | 2.0 | 0.0 | 3.0 | 0.0 | 0.0 | 19.44 | 1.0 | ... | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0 | 1 | 1 | 0 |
| ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... | ... |
| 2023-05-04 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 33.33 | 1.0 | ... | 0.0 | 63889986991.0 | 67171434888.0 | 181897083005.0 | 50835661126.0 | 1.0 | 0 | 1 | 0 | 1 |
| 2023-05-05 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 33.33 | 1.0 | ... | 0.0 | 63993367323.0 | 67276140121.0 | 182280782416.0 | 51011274972.0 | 1.0 | 0 | 1 | 0 | 1 |
| 2023-05-06 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 33.33 | 1.0 | ... | 0.0 | 64096748099.0 | 67380845596.0 | 182664482513.0 | 51186888818.0 | 1.0 | 0 | 1 | 0 | 1 |
| 2023-05-07 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 33.33 | 1.0 | ... | 0.0 | 64200128969.0 | 67485551123.0 | 183048182756.0 | 51362502664.0 | 1.0 | 0 | 1 | 0 | 1 |
| 2023-05-08 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 1.0 | 33.33 | 1.0 | ... | 0.0 | 64303509961.0 | 67590256732.0 | 183431912857.0 | 51538146164.0 | 1.0 | 0 | 1 | 0 | 1 |
1189 rows × 21 columns
2.1 Principal Component Analysis
To remove multicollinearity of indicators, we use pca package: a python package to perform Principal Component Analysis and to create insightful plots via our MLEngineer(seed=0).pca(X, n_components). Standardization (Z-score normalization) and Principal Component Analysis (PCA) will be performed.
[9]:
ml = cs.MLEngineer()
pca_dict = ml.pca(X=indicator_df, n_components=0.95)
pca_df = pca_dict["PC"].copy()
pca_df.tail()
[pca] WARNING use the standardized verbose status. The status [1-6] will be deprecated in future versions.
[9]:
| PC1 | PC2 | PC3 | PC4 | PC5 | PC6 | PC7 | PC8 | PC9 | PC10 | |
|---|---|---|---|---|---|---|---|---|---|---|
| 2023-05-04 | 5.673492 | -2.243826 | 0.555292 | -0.134307 | 0.888876 | 0.020971 | -0.364962 | -0.630145 | 0.039768 | -0.324168 |
| 2023-05-05 | 5.683267 | -2.247735 | 0.557146 | -0.134422 | 0.891652 | 0.019268 | -0.363681 | -0.632074 | 0.041499 | -0.325286 |
| 2023-05-06 | 5.693041 | -2.251644 | 0.558999 | -0.134538 | 0.894427 | 0.017564 | -0.362401 | -0.634004 | 0.043231 | -0.326404 |
| 2023-05-07 | 5.702815 | -2.255553 | 0.560853 | -0.134654 | 0.897203 | 0.015860 | -0.361120 | -0.635933 | 0.044962 | -0.327522 |
| 2023-05-08 | 5.712591 | -2.259462 | 0.562707 | -0.134770 | 0.899979 | 0.014157 | -0.359839 | -0.637862 | 0.046694 | -0.328641 |
The output of MLEngineer().pca() is the model of pca package, we can show some figures easily as follows.
Explained variance:
[10]:
pca_dict["model"].plot()
plt.close()
Top features:
[11]:
df = pca_dict["topfeat"].copy()
df["PC"] = df["PC"].str.extract(r"(\d+)").astype(np.int64)
df = df.sort_values(["PC", "type"]).reset_index(drop=True)
def highlight(d):
styles = d.copy()
styles.loc[:, :] = ""
styles.loc[d["type"] == "best", :] = "background-color: yellow"
return styles
df.style.apply(highlight, axis=None)
[11]:
| PC | feature | loading | type | |
|---|---|---|---|---|
| 0 | 1 | Vaccinated_once | 0.347669 | best |
| 1 | 1 | Contact_tracing | -0.294071 | weak |
| 2 | 1 | Information_campaigns | -0.281398 | weak |
| 3 | 1 | Vaccinated_full | 0.346382 | weak |
| 4 | 1 | Vaccinations | 0.343559 | weak |
| 5 | 1 | Vaccinations_boosters | 0.308128 | weak |
| 6 | 2 | Stringency_index | 0.438532 | best |
| 7 | 2 | International_movement_restrictions | 0.313789 | weak |
| 8 | 2 | Product_0 | -0.318462 | weak |
| 9 | 2 | Product_Moderna, Novavax, Oxford/AstraZeneca, Pfizer/BioNTech | 0.318462 | weak |
| 10 | 3 | Country_Japan | 0.677881 | best |
| 11 | 3 | Country_0 | -0.677881 | weak |
| 12 | 4 | School_closing | 0.558851 | best |
| 13 | 5 | Transport_closing | 0.554676 | best |
| 14 | 6 | Workplace_closing | 0.650936 | best |
| 15 | 7 | Tests | 0.570970 | best |
| 16 | 8 | Cancel_events | 0.513032 | best |
| 17 | 9 | Testing_policy | 0.664159 | best |
| 18 | 10 | Gatherings_restrictions | 0.525608 | best |
| 19 | 10 | Internal_movement_restrictions | 0.331885 | weak |
| 20 | 10 | Stay_home_restrictions | -0.421741 | weak |
2-2. Future values of indicators
Before prediction of ODE parameter values, we need to prepare future values of (PCA-performed) indicators. We can add future values to the pandas.DataFrame manually or forecast them with MLEngineer(seed=0).predict(Y, days=<int>, X=None) as follows.
[12]:
future_df = ml.forecast(Y=pca_df, days=30, X=None)
future_df.tail()
Using 1 cpus for n_jobs.
Data frequency is: D, used frequency is: D
Model Number: 1 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "SinTrend"}, "transformation_params": {"0": {}, "1": {}}}
Model Number: 2 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 3 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "Round", "2": "Detrend"}, "transformation_params": {"0": {"lag_1": 7, "method": "Mean"}, "1": {"model": "middle", "decimals": 2, "on_transform": true, "on_inverse": false}, "2": {"model": "GLS"}}}
Model Number: 4 with model GLS in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
Model Number: 5 with model GLS in generation 0 of 1 with params {} and transformations {"fillna": "median", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "RobustScaler", "3": "Round", "4": "MaxAbsScaler"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}, "3": {"model": "middle", "decimals": 2, "on_transform": true, "on_inverse": true}, "4": {}}}
Model Number: 6 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "bkfilter", "1": "SinTrend", "2": "Detrend", "3": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {}, "2": {"model": "Linear"}, "3": {}}}
Model Number: 7 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PositiveShift", "1": "SinTrend", "2": "bkfilter"}, "transformation_params": {"0": {}, "1": {}, "2": {}}}
Model Number: 8 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "SinTrend"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}}}
Model Number: 9 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 10 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 2, "lag_2": 7} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SinTrend", "1": "Round", "2": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {"model": "middle", "decimals": 2, "on_transform": false, "on_inverse": true}, "2": {}}}
Model Number: 11 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 2, "lag_2": 1} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "QuantileTransformer", "2": "Detrend"}, "transformation_params": {"0": {"lag_1": 12, "method": "Median"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"model": "GLS"}}}
Model Number: 12 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 7, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "QuantileTransformer", "1": "ClipOutliers"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 13 with model ConstantNaive in generation 0 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer", "2": "SeasonalDifference"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 7, "method": "LastValue"}}}
Model Number: 14 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 28} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 15 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "sokalmichener", "include_differenced": true, "k": 20, "stride_size": 1, "regression_type": null} and transformations {"fillna": "zero", "transformations": {"0": null}, "transformation_params": {"0": {}}}
Model Number: 16 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer", "2": "RobustScaler"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 17 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 30} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 18 with model SeasonalityMotif in generation 0 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "mae", "k": 10, "datepart_method": "common_fourier"} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
Model Number: 19 with model SectionalMotif in generation 0 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "AlignLastValue", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"lag_1": 7, "method": "Median"}, "1": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}}}
Model Number: 20 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "simple_binarized", "changepoint_spacing": 360, "changepoint_distance_end": 360, "regression_type": null, "lambda_": 0.01} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "Round", "1": "AlignLastValue", "2": "HistoricValues", "3": "ClipOutliers", "4": "bkfilter"}, "transformation_params": {"0": {"decimals": -1, "on_transform": true, "on_inverse": true}, "1": {"rows": 7, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": true, "threshold": null, "threshold_method": "mean"}, "2": {"window": 10}, "3": {"method": "clip", "std_threshold": 3, "fillna": null}, "4": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3627, in fit
np.linalg.inv(X_values.T @ X_values + self.lambda_ * I)
ValueError: matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 1159 is different from 2424)
in model 20 in generation 0: BasicLinearModel
Model Number: 21 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["dayofweek", [365.25, 14]], "changepoint_spacing": 90, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98} and transformations {"fillna": "piecewise_polynomial", "transformations": {"0": "AlignLastValue", "1": "IntermittentOccurrence", "2": "RobustScaler", "3": "Log"}, "transformation_params": {"0": {"rows": 1, "lag": 2, "method": "multiplicative", "strength": 0.9, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"center": "mean"}, "2": {}, "3": {}}}
Model Number: 22 with model SeasonalityMotif in generation 0 of 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
Model Number: 23 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "recurring", "changepoint_spacing": 28, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 0.01, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Slice", "1": "ClipOutliers", "2": "SeasonalDifference", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer", "5": "PositiveShift"}, "transformation_params": {"0": {"method": 0.2}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "5": {}}}
Model Number: 24 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "zero", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AnomalyRemoval"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": 30, "transform_dict": null}, "2": {"method": "zscore", "method_params": {"distribution": "uniform", "alpha": 0.14}, "fillna": "fake_date", "transform_dict": {"fillna": "quadratic", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 7, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}}}, "isolated_only": false, "on_inverse": false}}}
Model Number: 25 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastDiff", "1": "MaxAbsScaler", "2": "HolidayTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"rows": 28, "displacement_rows": 1, "quantile": 0.7, "decay_span": null}, "1": {}, "2": {"threshold": 0.8, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": false, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": true, "use_hindu_holidays": false, "anomaly_detector_params": {"method": "med_diff", "method_params": {"distribution": "beta", "alpha": 0.05}, "fillna": "ffill", "transform_dict": null, "isolated_only": true, "on_inverse": false}, "remove_excess_anomalies": false, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": [168, "hour"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {"rows": 1, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 26 with model AverageValueNaive in generation 0 of 1 with params {"method": "Midhinge", "window": null} and transformations {"fillna": "median", "transformations": {"0": "ScipyFilter", "1": "HPFilter", "2": "AlignLastValue", "3": "UpscaleDownscaleTransformer"}, "transformation_params": {"0": {"method": "savgol_filter", "method_args": {"window_length": 31, "polyorder": 1, "deriv": 0, "mode": "mirror"}}, "1": {"part": "trend", "lamb": 129600}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "3": {"mode": "upscale", "factor": 1, "down_method": "decimate", "fill_method": "linear"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2288, in TemplateWizard
template_result, best_smape = _eval_prediction_for_template(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1637, in _eval_prediction_for_template
model_error = df_forecast.evaluate(
^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/base.py", line 830, in evaluate
) = full_metric_evaluation(
^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/metrics.py", line 630, in full_metric_evaluation
full_errors = F - A
~~^~~
ValueError: operands could not be broadcast together with shapes (60,10) (30,10)
in model 26 in generation 0: AverageValueNaive
Model Number: 27 with model GLS in generation 0 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "ChangepointDetrend"}, "transformation_params": {"0": {"model": "Linear", "changepoint_spacing": 5040, "changepoint_distance_end": 28, "datepart_method": null}}}
Model Number: 28 with model SeasonalNaive in generation 0 of 1 with params {"method": "mean", "lag_1": 2, "lag_2": 60} and transformations {"fillna": "rolling_mean", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}}}
Model Number: 29 with model SeasonalityMotif in generation 0 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 30 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "yule", "include_differenced": true, "k": 3, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "EWMAFilter"}, "transformation_params": {"0": {"rows": null, "displacement_rows": 4, "quantile": 1.0, "decay_span": null}, "1": {"span": 12}}}, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "quadratic", "transformations": {"0": "AlignLastValue", "1": "PCA", "2": "QuantileTransformer", "3": "AlignLastValue", "4": "HPFilter", "5": "CenterSplit"}, "transformation_params": {"0": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "1": {"whiten": false, "n_components": 0.3}, "2": {"output_distribution": "uniform", "n_quantiles": 1000}, "3": {"rows": 168, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "max"}, "4": {"part": "trend", "lamb": 1600}, "5": {"fillna": "ffill", "center": "median"}}}
Model Number: 31 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill", "transformations": {"0": "LevelShiftTransformer", "1": "cffilter", "2": "MaxAbsScaler", "3": "SinTrend"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}, "1": {}, "2": {}, "3": {}}}
Model Number: 32 with model SeasonalityMotif in generation 0 of 1 with params {"window": 30, "point_method": "mean", "distance_metric": "mae", "k": 5, "datepart_method": "recurring", "independent": false} and transformations {"fillna": "median", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "RegressionFilter", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 5, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "2": {"sigma": 1, "rolling_window": 90, "run_order": "season_first", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": null, "min_samples_split": 1.0}}, "datepart_method": "expanded_binarized", "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "holiday_params": null, "trend_method": "rolling_mean"}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 33 with model AverageValueNaive in generation 0 of 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": null, "min_samples_split": 1.0}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
Model Number: 34 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 35 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 90, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "IntermittentOccurrence", "2": "Discretize", "3": "Discretize", "4": "AlignLastValue", "5": "RollingMeanTransformer"}, "transformation_params": {"0": {"rows": 28, "displacement_rows": 1, "quantile": 1.0, "decay_span": 365}, "1": {"center": "mean"}, "2": {"discretization": "upper", "n_bins": 10}, "3": {"discretization": "center", "n_bins": 20}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "5": {"fixed": true, "window": 3, "macro_micro": false, "center": true}}}
Model Number: 36 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 37 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "median", "distance_metric": "jaccard", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "rolling_mean", "transformations": {"0": "QuantileTransformer", "1": "AlignLastValue", "2": "StandardScaler", "3": "CumSumTransformer"}, "transformation_params": {"0": {"output_distribution": "normal", "n_quantiles": "fifth"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.7, "first_value_only": false, "threshold": 3, "threshold_method": "max"}, "2": {}, "3": {}}}
Model Number: 38 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 90, "changepoint_distance_end": null, "regression_type": null, "lambda_": 0.01, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "IntermittentOccurrence", "1": "AlignLastValue", "2": "AlignLastValue", "3": "RollingMeanTransformer", "4": "AlignLastValue", "5": "AlignLastValue"}, "transformation_params": {"0": {"center": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "3": {"fixed": false, "window": 12, "macro_micro": true, "center": false}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "5": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 39 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
Model Number: 40 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 41 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "akima", "transformations": {"0": "MinMaxScaler", "1": "SeasonalDifference", "2": "MinMaxScaler", "3": "EWMAFilter", "4": "LevelShiftTransformer", "5": "Detrend"}, "transformation_params": {"0": {}, "1": {"lag_1": 12, "method": "LastValue"}, "2": {}, "3": {"span": 10}, "4": {"window_size": 14, "alpha": 4.0, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "rolling_diff_5nn"}, "5": {"model": "Linear", "phi": 1, "window": null, "transform_dict": null}}}
Model Number: 42 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 70, "lag_2": null} and transformations {"fillna": "pchip", "transformations": {"0": "CenterLastValue", "1": "DifferencedTransformer", "2": "Detrend"}, "transformation_params": {"0": {"rows": 3}, "1": {"lag": 1, "fill": "bfill"}, "2": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}}}
Model Number: 43 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 180, "constant": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue", "4": "Round", "5": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"decimals": 0, "on_transform": false, "on_inverse": true}, "5": {"lag_1": 7, "method": "Median"}}}
Model Number: 44 with model GLS in generation 0 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}}}
Model Number: 45 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "quadratic", "transformations": {"0": "PctChangeTransformer", "1": "MeanPercentSplitter", "2": "StandardScaler"}, "transformation_params": {"0": {}, "1": {"window": 3}, "2": {}}}
Model Number: 46 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 47 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
Model Number: 48 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 5040, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 0.001, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}}}
Model Number: 49 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 180, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "RobustScaler", "1": "Discretize", "2": "CenterSplit"}, "transformation_params": {"0": {}, "1": {"discretization": "center", "n_bins": 5}, "2": {"fillna": "mean", "center": "zero"}}}
Model Number: 50 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["day", 168, 52], "changepoint_spacing": 90, "changepoint_distance_end": 90, "regression_type": "User", "lambda_": null, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.5, "first_value_only": false, "threshold": 10, "threshold_method": "max"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3605, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 50 in generation 0: BasicLinearModel
Model Number: 51 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "minkowski", "include_differenced": true, "k": 100, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue", "4": "AlignLastValue", "5": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"model": "Linear", "phi": 0.999, "window": 90, "transform_dict": {"fillna": "rolling_mean_24", "transformations": {"0": "RollingMean100thN", "1": "DifferencedTransformer"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "bfill"}}}}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "5": {"rows": 2}}}
Model Number: 52 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "cityblock", "include_differenced": true, "k": 15, "stride_size": 1, "regression_type": "User", "comparison_transformation": {"fillna": "fake_date", "transformations": {"0": "Detrend", "1": "DifferencedTransformer"}, "transformation_params": {"0": {"model": "GLS", "phi": 0.999, "window": null, "transform_dict": null}, "1": {"lag": 1, "fill": "bfill"}}}, "combination_transformation": null} and transformations {"fillna": "time", "transformations": {"0": "ClipOutliers", "1": "Detrend"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": {"fillna": null, "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4}}}}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 1847, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 52 in generation 0: SectionalMotif
Model Number: 53 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "expanded_binarized", "changepoint_spacing": 180, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 0.1, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "DatepartRegression", "1": "AlignLastValue", "2": "SinTrend", "3": "MinMaxScaler", "4": "AlignLastValue"}, "transformation_params": {"0": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 5000}}, "datepart_method": [7, 365.25], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "1": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3627, in fit
np.linalg.inv(X_values.T @ X_values + self.lambda_ * I)
ValueError: matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 1159 is different from 2424)
in model 53 in generation 0: BasicLinearModel
Model Number: 54 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": null, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "STLFilter", "1": "MinMaxScaler", "2": "StandardScaler"}, "transformation_params": {"0": {"decomp_type": "STL", "part": "trend", "seasonal": 7}, "1": {}, "2": {}}}
Model Number: 55 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "pchip", "transformations": {"0": "AlignLastValue", "1": "ThetaTransformer", "2": "DifferencedTransformer", "3": "StandardScaler", "4": "DifferencedTransformer"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "max"}, "1": {"theta_values": [0.8, 1.2]}, "2": {"lag": 1, "fill": "bfill"}, "3": {}, "4": {"lag": 1, "fill": "zero"}}}
Model Number: 56 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "pchip", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "DifferencedTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 365, "transform_dict": null}, "2": {"lag": 1, "fill": "bfill"}, "3": {"rows": 168, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 57 with model AverageValueNaive in generation 0 of 1 with params {"method": "Median", "window": null} and transformations {"fillna": "ffill", "transformations": {"0": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": "quarter"}}}
Model Number: 58 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["morlet_365.25_12_12", "ricker_7_7_1"], "changepoint_spacing": 60, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Detrend", "1": "IntermittentOccurrence", "2": "MeanPercentSplitter"}, "transformation_params": {"0": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "1": {"center": "mean"}, "2": {"window": "forecast_length"}}}
Model Number: 59 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "midhinge", "distance_metric": "correlation", "include_differenced": true, "k": 1, "stride_size": 1, "regression_type": "User", "comparison_transformation": {"fillna": "quadratic", "transformations": {"0": "Log"}, "transformation_params": {"0": {}}}, "combination_transformation": null} and transformations {"fillna": "time", "transformations": {"0": "FFTFilter"}, "transformation_params": {"0": {"cutoff": 0.6, "reverse": false, "on_transform": true, "on_inverse": false}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 1847, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 59 in generation 0: SectionalMotif
Model Number: 60 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 5040, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "quadratic", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 61 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "zero", "transformations": {"0": "AnomalyRemoval", "1": "Constraint", "2": "MinMaxScaler"}, "transformation_params": {"0": {"method": "minmax", "method_params": {"alpha": 0.03}, "fillna": "ffill", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HistoricValues", "1": "PowerTransformer"}, "transformation_params": {"0": {"window": null}, "1": {}}}, "isolated_only": false, "on_inverse": false}, "1": {"constraint_method": "stdev_min", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 3.0, "bounds_only": false, "fillna": "ffill"}, "2": {}}}
Model Number: 62 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 180, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 100, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "pchip", "transformations": {"0": "STLFilter", "1": "AlignLastValue"}, "transformation_params": {"0": {"decomp_type": "seasonal_decompose", "part": "trend"}, "1": {"rows": 84, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 63 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 60, "changepoint_distance_end": 520, "regression_type": null, "lambda_": 1, "trend_phi": 0.97, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}}}
New Generation: 1 of 1
Model Number: 64 with model SeasonalityMotif in generation 1 of 1 with params {"window": 7, "point_method": "mean", "distance_metric": "canberra", "k": 5, "datepart_method": "expanded", "independent": false} and transformations {"fillna": "cubic", "transformations": {"0": "Discretize", "1": "DifferencedTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"discretization": "upper", "n_bins": 20}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}
Model Number: 65 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "zero", "transformations": {"0": "ThetaTransformer", "1": "QuantileTransformer", "2": "SeasonalDifference"}, "transformation_params": {"0": {"theta_values": [0.6, 1.4]}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 7, "method": "LastValue"}}}
Model Number: 66 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 28} and transformations {"fillna": "ffill", "transformations": {"0": "MinMaxScaler", "1": "CenterSplit"}, "transformation_params": {"0": {}, "1": {"fillna": "akima", "center": "zero"}}}
Model Number: 67 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "ThetaTransformer"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"theta_values": [0, 2]}}}
Model Number: 68 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "ScipyFilter", "3": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {}, "2": {"method": "savgol_filter", "method_args": {"window_length": 7, "polyorder": 4, "deriv": 0, "mode": "nearest"}}, "3": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 69 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "ClipOutliers", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 70 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "StandardScaler", "1": "RobustScaler", "2": "Round", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 71 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": null}, "transformation_params": {"0": {}}}
Model Number: 72 with model SectionalMotif in generation 1 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "canberra", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "bkfilter", "2": "Discretize", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {}, "2": {"discretization": "center", "n_bins": 20}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 73 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "RobustScaler"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}}}
Model Number: 74 with model GLS in generation 1 of 1 with params {"changepoint_spacing": 5040, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}}}
Model Number: 75 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "QuantileTransformer", "2": "PctChangeTransformer"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 76 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 60, "changepoint_distance_end": 520, "regression_type": null, "lambda_": 1, "trend_phi": 0.97, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "RegressionFilter", "1": "QuantileTransformer", "2": "Detrend", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer"}, "transformation_params": {"0": {"sigma": 1, "rolling_window": 90, "run_order": "season_first", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 9, "min_samples_split": 2}}, "datepart_method": "simple_binarized", "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}, "holiday_params": null, "trend_method": "rolling_mean"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"model": "Linear", "phi": 1, "window": null, "transform_dict": null}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}}}
Model Number: 77 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "weighted_mean", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "cubic", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue", "4": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 2}}}
Model Number: 78 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "AlignLastValue", "3": "PositiveShift"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "3": {}}}
Model Number: 79 with model SeasonalityMotif in generation 1 of 1 with params {"window": 7, "point_method": "mean", "distance_metric": "mae", "k": 1, "datepart_method": "recurring", "independent": true} and transformations {"fillna": "linear", "transformations": {"0": "Discretize", "1": "AlignLastValue", "2": "AlignLastValue", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"discretization": "upper", "n_bins": 20}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 80 with model SeasonalityMotif in generation 1 of 1 with params {"window": 30, "point_method": "trimmed_mean_40", "distance_metric": "minkowski", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": false} and transformations {"fillna": "pchip", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}
Model Number: 81 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 2, "lag_2": 7} and transformations {"fillna": "fake_date", "transformations": {"0": "Round", "1": "Round", "2": "ThetaTransformer"}, "transformation_params": {"0": {"decimals": 1, "on_transform": true, "on_inverse": false}, "1": {"model": "middle", "decimals": 2, "on_transform": false, "on_inverse": true}, "2": {"theta_values": [0.6, 1.4]}}}
Model Number: 82 with model AverageValueNaive in generation 1 of 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "ScipyFilter", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {}, "2": {"method": "savgol_filter", "method_args": {"window_length": 7, "polyorder": 4, "deriv": 0, "mode": "nearest"}}, "3": {}, "4": {}}}
Model Number: 83 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "FFTFilter", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {"cutoff": 0.01, "reverse": false, "on_transform": true, "on_inverse": false}, "1": {"lag": 1, "fill": "bfill"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 84 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 2, "lag_2": 2} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "MeanPercentSplitter", "1": "QuantileTransformer"}, "transformation_params": {"0": {"window": 10}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 85 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "bkfilter", "1": "AlignLastValue", "2": "LevelShiftTransformer"}, "transformation_params": {"0": {}, "1": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {"window_size": 90, "alpha": 3.5, "grouping_forward_limit": 5, "max_level_shifts": 8, "alignment": "rolling_diff"}}}
Model Number: 86 with model SectionalMotif in generation 1 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "sokalmichener", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"lag_1": 7, "method": "Median"}, "1": {"lag": 1, "fill": "zero"}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}}}
Model Number: 87 with model ConstantNaive in generation 1 of 1 with params {"constant": -1} and transformations {"fillna": "cubic", "transformations": {"0": "ThetaTransformer", "1": "ClipOutliers", "2": "PctChangeTransformer"}, "transformation_params": {"0": {"theta_values": [0.6, 1.4]}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {}}}
Model Number: 88 with model SeasonalNaive in generation 1 of 1 with params {"method": "mean", "lag_1": 10, "lag_2": 96} and transformations {"fillna": "cubic", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer", "2": "AnomalyRemoval", "3": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": null, "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 6}}}, "isolated_only": true, "on_inverse": false}, "3": {"output_distribution": "uniform", "n_quantiles": "quarter"}}}
Model Number: 89 with model SeasonalNaive in generation 1 of 1 with params {"method": "mean", "lag_1": 28, "lag_2": 1} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "PctChangeTransformer", "2": "Detrend", "3": "QuantileTransformer"}, "transformation_params": {"0": {"lag_1": 12, "method": "Median"}, "1": {}, "2": {"model": "GLS"}, "3": {"output_distribution": "normal", "n_quantiles": 1000}}}
Model Number: 90 with model SectionalMotif in generation 1 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "AlignLastValue", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 91 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "DifferencedTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {}}}
Model Number: 92 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 53, "lag_2": 28} and transformations {"fillna": "time", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}}}
Model Number: 93 with model SeasonalityMotif in generation 1 of 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "minkowski", "k": 5, "datepart_method": "simple_2", "independent": true} and transformations {"fillna": "pchip", "transformations": {"0": "Round", "1": "RollingMeanTransformer", "2": "StandardScaler", "3": "FIRFilter"}, "transformation_params": {"0": {"decimals": 2, "on_transform": true, "on_inverse": true}, "1": {"fixed": true, "window": 7, "macro_micro": false, "center": true}, "2": {}, "3": {"numtaps": 128, "cutoff_hz": 0.1, "window": "boxcar", "sampling_frequency": 28, "on_transform": true, "on_inverse": false, "bounds_only": false}}}
Model Number: 94 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 95 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "Round", "4": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"decimals": 0, "on_transform": false, "on_inverse": true}, "4": {"lag_1": 7, "method": "Median"}}}
Model Number: 96 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 60, "changepoint_distance_end": 520, "regression_type": null, "lambda_": 1, "trend_phi": 0.97, "holiday_countries_used": true} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "ClipOutliers", "2": "PctChangeTransformer", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}}}
Model Number: 97 with model SectionalMotif in generation 1 of 1 with params {"window": 5, "point_method": "median", "distance_metric": "minkowski", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "ScipyFilter", "1": "DatepartRegression", "2": "AlignLastValue"}, "transformation_params": {"0": {"method": "butter", "method_args": {"N": 3, "window_size": 2, "btype": "lowpass", "analog": false, "output": "sos"}}, "1": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": "simple", "polynomial_degree": null, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}, "holiday_countries_used": false, "lags": null, "forward_lags": 2}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 98 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "recurring", "changepoint_spacing": 180, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 0.01, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "piecewise_polynomial", "transformations": {"0": "RollingMeanTransformer", "1": "AlignLastValue", "2": "AlignLastValue", "3": "Log"}, "transformation_params": {"0": {"fixed": true, "window": 3, "macro_micro": false, "center": false}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "3": {}}}
Model Number: 99 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 180, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Slice", "1": "ClipOutliers", "2": "SeasonalDifference", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer"}, "transformation_params": {"0": {"method": 0.2}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}}}
Model Number: 100 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "wasserstein", "include_differenced": true, "k": 15, "stride_size": 5, "regression_type": "User", "comparison_transformation": {"fillna": "zero", "transformations": {"0": "PositiveShift"}, "transformation_params": {"0": {}}}, "combination_transformation": {"fillna": "ffill_mean_biased", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}} and transformations {"fillna": "zero", "transformations": {"0": "QuantileTransformer", "1": "HPFilter"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"part": "trend", "lamb": 1600}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 1847, in fit
raise ValueError(
ValueError: regression_type=='User' but no future_regressor supplied
in model 100 in generation 1: SectionalMotif
Model Number: 101 with model SeasonalityMotif in generation 1 of 1 with params {"window": 50, "point_method": "trimmed_mean_40", "distance_metric": "mse", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 102 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "rolling_mean", "transformations": {"0": "bkfilter", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue", "4": "ChangepointDetrend", "5": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"model": "Linear", "changepoint_spacing": null, "changepoint_distance_end": 5040, "datepart_method": "common_fourier_rw"}, "5": {"rows": 4, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "max"}}}
Model Number: 103 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 2, "lag_2": 7} and transformations {"fillna": "cubic", "transformations": {"0": "SinTrend", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Validation Round: 1
Validation train index is DatetimeIndex(['2020-02-05', '2020-02-06', '2020-02-07', '2020-02-08',
'2020-02-09', '2020-02-10', '2020-02-11', '2020-02-12',
'2020-02-13', '2020-02-14',
...
'2023-02-28', '2023-03-01', '2023-03-02', '2023-03-03',
'2023-03-04', '2023-03-05', '2023-03-06', '2023-03-07',
'2023-03-08', '2023-03-09'],
dtype='datetime64[ns]', length=1129, freq=None)
Model Number: 1 of 16 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
📈 1 - ConstantNaive with avg smape 120.0:
Model Number: 2 of 16 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "ClipOutliers", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
2 - ConstantNaive with avg smape 120.01:
Model Number: 3 of 16 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
3 - ConstantNaive with avg smape 120.03:
Model Number: 4 of 16 with model SectionalMotif for Validation 1 with params {"window": 5, "point_method": "median", "distance_metric": "minkowski", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "ScipyFilter", "1": "DatepartRegression", "2": "AlignLastValue"}, "transformation_params": {"0": {"method": "butter", "method_args": {"N": 3, "btype": "lowpass", "analog": false, "output": "sos", "Wn": 0.5}}, "1": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": "simple", "polynomial_degree": null, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}, "holiday_countries_used": false, "lags": null, "forward_lags": 2}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
4 - SectionalMotif with avg smape 120.74:
Model Number: 5 of 16 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "Round", "4": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"decimals": 0, "on_transform": false, "on_inverse": true}, "4": {"lag_1": 7, "method": "Median"}}}
5 - LastValueNaive with avg smape 131.55:
Model Number: 6 of 16 with model BasicLinearModel for Validation 1 with params {"datepart_method": "recurring", "changepoint_spacing": 28, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 0.01, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Slice", "1": "ClipOutliers", "2": "SeasonalDifference", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer", "5": "PositiveShift"}, "transformation_params": {"0": {"method": 0.2}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "5": {}}}
6 - BasicLinearModel with avg smape 123.89:
Model Number: 7 of 16 with model SectionalMotif for Validation 1 with params {"window": 7, "point_method": "weighted_mean", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "cubic", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue", "4": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 2}}}
7 - SectionalMotif with avg smape 121.9:
Model Number: 8 of 16 with model AverageValueNaive for Validation 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
8 - AverageValueNaive with avg smape 123.67:
Model Number: 9 of 16 with model SectionalMotif for Validation 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "AlignLastValue", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"lag_1": 7, "method": "Median"}, "1": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}}}
9 - SectionalMotif with avg smape 143.43:
Model Number: 10 of 16 with model GLS for Validation 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
10 - GLS with avg smape 139.78:
Model Number: 11 of 16 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
11 - LastValueNaive with avg smape 121.61:
Model Number: 12 of 16 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
12 - LastValueNaive with avg smape 124.33:
Model Number: 13 of 16 with model SeasonalityMotif for Validation 1 with params {"window": 50, "point_method": "trimmed_mean_40", "distance_metric": "mse", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
13 - SeasonalityMotif with avg smape 124.59:
Model Number: 14 of 16 with model SeasonalityMotif for Validation 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
14 - SeasonalityMotif with avg smape 124.57:
Model Number: 15 of 16 with model SeasonalityMotif for Validation 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
15 - SeasonalityMotif with avg smape 125.29:
Model Number: 16 of 16 with model AverageValueNaive for Validation 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
16 - AverageValueNaive with avg smape 134.34:
Validation Round: 2
Validation train index is DatetimeIndex(['2020-02-05', '2020-02-06', '2020-02-07', '2020-02-08',
'2020-02-09', '2020-02-10', '2020-02-11', '2020-02-12',
'2020-02-13', '2020-02-14',
...
'2023-01-29', '2023-01-30', '2023-01-31', '2023-02-01',
'2023-02-02', '2023-02-03', '2023-02-04', '2023-02-05',
'2023-02-06', '2023-02-07'],
dtype='datetime64[ns]', length=1099, freq=None)
Model Number: 1 of 16 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
📈 1 - ConstantNaive with avg smape 120.04:
Model Number: 2 of 16 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "ClipOutliers", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
📈 2 - ConstantNaive with avg smape 120.02:
Model Number: 3 of 16 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
3 - ConstantNaive with avg smape 120.18:
Model Number: 4 of 16 with model SectionalMotif for Validation 2 with params {"window": 5, "point_method": "median", "distance_metric": "minkowski", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "ScipyFilter", "1": "DatepartRegression", "2": "AlignLastValue"}, "transformation_params": {"0": {"method": "butter", "method_args": {"N": 3, "btype": "lowpass", "analog": false, "output": "sos", "Wn": 0.5}}, "1": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": "simple", "polynomial_degree": null, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}, "holiday_countries_used": false, "lags": null, "forward_lags": 2}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
4 - SectionalMotif with avg smape 121.2:
Model Number: 5 of 16 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "Round", "4": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"decimals": 0, "on_transform": false, "on_inverse": true}, "4": {"lag_1": 7, "method": "Median"}}}
5 - LastValueNaive with avg smape 129.74:
Model Number: 6 of 16 with model BasicLinearModel for Validation 2 with params {"datepart_method": "recurring", "changepoint_spacing": 28, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 0.01, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Slice", "1": "ClipOutliers", "2": "SeasonalDifference", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer", "5": "PositiveShift"}, "transformation_params": {"0": {"method": 0.2}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "5": {}}}
6 - BasicLinearModel with avg smape 123.39:
Model Number: 7 of 16 with model SectionalMotif for Validation 2 with params {"window": 7, "point_method": "weighted_mean", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "cubic", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue", "4": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "zero"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 2}}}
7 - SectionalMotif with avg smape 121.5:
Model Number: 8 of 16 with model AverageValueNaive for Validation 2 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "QuantileTransformer"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
8 - AverageValueNaive with avg smape 123.11:
Model Number: 9 of 16 with model SectionalMotif for Validation 2 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "AlignLastValue", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"lag_1": 7, "method": "Median"}, "1": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}}}
9 - SectionalMotif with avg smape 121.79:
Model Number: 10 of 16 with model GLS for Validation 2 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
10 - GLS with avg smape 139.65:
Model Number: 11 of 16 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
11 - LastValueNaive with avg smape 122.6:
Model Number: 12 of 16 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
12 - LastValueNaive with avg smape 123.55:
Model Number: 13 of 16 with model SeasonalityMotif for Validation 2 with params {"window": 50, "point_method": "trimmed_mean_40", "distance_metric": "mse", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
13 - SeasonalityMotif with avg smape 123.77:
Model Number: 14 of 16 with model SeasonalityMotif for Validation 2 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
14 - SeasonalityMotif with avg smape 123.84:
Model Number: 15 of 16 with model SeasonalityMotif for Validation 2 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
15 - SeasonalityMotif with avg smape 124.4:
Model Number: 16 of 16 with model AverageValueNaive for Validation 2 with params {"method": "Mean", "window": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
16 - AverageValueNaive with avg smape 122.24:
Validation Round: 0
Validation train index is DatetimeIndex(['2020-02-05', '2020-02-06', '2020-02-07', '2020-02-08',
'2020-02-09', '2020-02-10', '2020-02-11', '2020-02-12',
'2020-02-13', '2020-02-14',
...
'2023-03-27', '2023-03-28', '2023-03-29', '2023-03-30',
'2023-03-31', '2023-04-01', '2023-04-02', '2023-04-03',
'2023-04-04', '2023-04-05'],
dtype='datetime64[ns]', length=1156, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 1 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 3, "model_metric": "Score", "models": {"9038af4be1fa5c92d10f5baa836ac744": {"Model": "ConstantNaive", "ModelParameters": "{\"constant\": 0}", "TransformationParameters": "{\"fillna\": \"cubic\", \"transformations\": {\"0\": \"StandardScaler\", \"1\": \"DifferencedTransformer\", \"2\": \"PctChangeTransformer\", \"3\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {}, \"1\": {\"lag\": 1, \"fill\": \"zero\"}, \"2\": {}, \"3\": {\"rows\": 1, \"lag\": 2, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}}}"}, "f8a2bc656be3e6fc565eed41cce958f6": {"Model": "ConstantNaive", "ModelParameters": "{\"constant\": 0}", "TransformationParameters": "{\"fillna\": \"cubic\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"DifferencedTransformer\", \"2\": \"AlignLastValue\", \"3\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 3.5, \"fillna\": null}, \"1\": {\"lag\": 1, \"fill\": \"zero\"}, \"2\": {\"rows\": 4, \"lag\": 2, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}, \"3\": {\"rows\": 1, \"lag\": 7, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 1, \"threshold_method\": \"mean\"}}}"}, "0145a925ee73e6ded975aa2f49c1e502": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"fake_date\", \"transformations\": {\"0\": \"SeasonalDifference\", \"1\": \"DifferencedTransformer\", \"2\": \"AlignLastValue\", \"3\": \"Round\", \"4\": \"SeasonalDifference\"}, \"transformation_params\": {\"0\": {\"lag_1\": 7, \"method\": \"LastValue\"}, \"1\": {\"lag\": 1, \"fill\": \"bfill\"}, \"2\": {\"rows\": 84, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 1, \"threshold_method\": \"max\"}, \"3\": {\"decimals\": 0, \"on_transform\": false, \"on_inverse\": true}, \"4\": {\"lag_1\": 7, \"method\": \"Median\"}}}"}}, "series": {"PC1": "f8a2bc656be3e6fc565eed41cce958f6", "PC2": "0145a925ee73e6ded975aa2f49c1e502", "PC3": "f8a2bc656be3e6fc565eed41cce958f6", "PC4": "0145a925ee73e6ded975aa2f49c1e502", "PC5": "f8a2bc656be3e6fc565eed41cce958f6", "PC6": "f8a2bc656be3e6fc565eed41cce958f6", "PC7": "0145a925ee73e6ded975aa2f49c1e502", "PC8": "0145a925ee73e6ded975aa2f49c1e502", "PC9": "9038af4be1fa5c92d10f5baa836ac744", "PC10": "0145a925ee73e6ded975aa2f49c1e502"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 3 ConstantNaive started
Ensemble Horizontal component 2 of 3 ConstantNaive started
Ensemble Horizontal component 3 of 3 LastValueNaive started
📈 1 - Ensemble with avg smape 104.01:
2 - Ensemble with avg smape 120.02:
📈 3 - Ensemble with avg smape 20.54:
4 - Ensemble with avg smape 24.39:
5 - Ensemble with avg smape 36.22:
6 - Ensemble with avg smape 26.56:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Validation Round: 1
Validation train index is DatetimeIndex(['2020-02-05', '2020-02-06', '2020-02-07', '2020-02-08',
'2020-02-09', '2020-02-10', '2020-02-11', '2020-02-12',
'2020-02-13', '2020-02-14',
...
'2023-02-26', '2023-02-27', '2023-02-28', '2023-03-01',
'2023-03-02', '2023-03-03', '2023-03-04', '2023-03-05',
'2023-03-06', '2023-03-07'],
dtype='datetime64[ns]', length=1127, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 2 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 3, "model_metric": "Score", "models": {"9038af4be1fa5c92d10f5baa836ac744": {"Model": "ConstantNaive", "ModelParameters": "{\"constant\": 0}", "TransformationParameters": "{\"fillna\": \"cubic\", \"transformations\": {\"0\": \"StandardScaler\", \"1\": \"DifferencedTransformer\", \"2\": \"PctChangeTransformer\", \"3\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {}, \"1\": {\"lag\": 1, \"fill\": \"zero\"}, \"2\": {}, \"3\": {\"rows\": 1, \"lag\": 2, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}}}"}, "f8a2bc656be3e6fc565eed41cce958f6": {"Model": "ConstantNaive", "ModelParameters": "{\"constant\": 0}", "TransformationParameters": "{\"fillna\": \"cubic\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"DifferencedTransformer\", \"2\": \"AlignLastValue\", \"3\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 3.5, \"fillna\": null}, \"1\": {\"lag\": 1, \"fill\": \"zero\"}, \"2\": {\"rows\": 4, \"lag\": 2, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}, \"3\": {\"rows\": 1, \"lag\": 7, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 1, \"threshold_method\": \"mean\"}}}"}, "0145a925ee73e6ded975aa2f49c1e502": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"fake_date\", \"transformations\": {\"0\": \"SeasonalDifference\", \"1\": \"DifferencedTransformer\", \"2\": \"AlignLastValue\", \"3\": \"Round\", \"4\": \"SeasonalDifference\"}, \"transformation_params\": {\"0\": {\"lag_1\": 7, \"method\": \"LastValue\"}, \"1\": {\"lag\": 1, \"fill\": \"bfill\"}, \"2\": {\"rows\": 84, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 1, \"threshold_method\": \"max\"}, \"3\": {\"decimals\": 0, \"on_transform\": false, \"on_inverse\": true}, \"4\": {\"lag_1\": 7, \"method\": \"Median\"}}}"}}, "series": {"PC1": "f8a2bc656be3e6fc565eed41cce958f6", "PC2": "0145a925ee73e6ded975aa2f49c1e502", "PC3": "f8a2bc656be3e6fc565eed41cce958f6", "PC4": "0145a925ee73e6ded975aa2f49c1e502", "PC5": "f8a2bc656be3e6fc565eed41cce958f6", "PC6": "f8a2bc656be3e6fc565eed41cce958f6", "PC7": "0145a925ee73e6ded975aa2f49c1e502", "PC8": "0145a925ee73e6ded975aa2f49c1e502", "PC9": "9038af4be1fa5c92d10f5baa836ac744", "PC10": "0145a925ee73e6ded975aa2f49c1e502"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 3 ConstantNaive started
Ensemble Horizontal component 2 of 3 ConstantNaive started
Ensemble Horizontal component 3 of 3 LastValueNaive started
📈 1 - Ensemble with avg smape 120.01:
2 - Ensemble with avg smape 120.02:
📈 3 - Ensemble with avg smape 12.29:
4 - Ensemble with avg smape 78.13:
5 - Ensemble with avg smape 21.27:
6 - Ensemble with avg smape 13.49:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Validation Round: 2
Validation train index is DatetimeIndex(['2020-02-05', '2020-02-06', '2020-02-07', '2020-02-08',
'2020-02-09', '2020-02-10', '2020-02-11', '2020-02-12',
'2020-02-13', '2020-02-14',
...
'2023-01-28', '2023-01-29', '2023-01-30', '2023-01-31',
'2023-02-01', '2023-02-02', '2023-02-03', '2023-02-04',
'2023-02-05', '2023-02-06'],
dtype='datetime64[ns]', length=1098, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 3 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 3, "model_metric": "Score", "models": {"9038af4be1fa5c92d10f5baa836ac744": {"Model": "ConstantNaive", "ModelParameters": "{\"constant\": 0}", "TransformationParameters": "{\"fillna\": \"cubic\", \"transformations\": {\"0\": \"StandardScaler\", \"1\": \"DifferencedTransformer\", \"2\": \"PctChangeTransformer\", \"3\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {}, \"1\": {\"lag\": 1, \"fill\": \"zero\"}, \"2\": {}, \"3\": {\"rows\": 1, \"lag\": 2, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}}}"}, "f8a2bc656be3e6fc565eed41cce958f6": {"Model": "ConstantNaive", "ModelParameters": "{\"constant\": 0}", "TransformationParameters": "{\"fillna\": \"cubic\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"DifferencedTransformer\", \"2\": \"AlignLastValue\", \"3\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 3.5, \"fillna\": null}, \"1\": {\"lag\": 1, \"fill\": \"zero\"}, \"2\": {\"rows\": 4, \"lag\": 2, \"method\": \"multiplicative\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}, \"3\": {\"rows\": 1, \"lag\": 7, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 1, \"threshold_method\": \"mean\"}}}"}, "0145a925ee73e6ded975aa2f49c1e502": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"fake_date\", \"transformations\": {\"0\": \"SeasonalDifference\", \"1\": \"DifferencedTransformer\", \"2\": \"AlignLastValue\", \"3\": \"Round\", \"4\": \"SeasonalDifference\"}, \"transformation_params\": {\"0\": {\"lag_1\": 7, \"method\": \"LastValue\"}, \"1\": {\"lag\": 1, \"fill\": \"bfill\"}, \"2\": {\"rows\": 84, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": 1, \"threshold_method\": \"max\"}, \"3\": {\"decimals\": 0, \"on_transform\": false, \"on_inverse\": true}, \"4\": {\"lag_1\": 7, \"method\": \"Median\"}}}"}}, "series": {"PC1": "f8a2bc656be3e6fc565eed41cce958f6", "PC2": "0145a925ee73e6ded975aa2f49c1e502", "PC3": "f8a2bc656be3e6fc565eed41cce958f6", "PC4": "0145a925ee73e6ded975aa2f49c1e502", "PC5": "f8a2bc656be3e6fc565eed41cce958f6", "PC6": "f8a2bc656be3e6fc565eed41cce958f6", "PC7": "0145a925ee73e6ded975aa2f49c1e502", "PC8": "0145a925ee73e6ded975aa2f49c1e502", "PC9": "9038af4be1fa5c92d10f5baa836ac744", "PC10": "0145a925ee73e6ded975aa2f49c1e502"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 3 ConstantNaive started
Ensemble Horizontal component 2 of 3 ConstantNaive started
Ensemble Horizontal component 3 of 3 LastValueNaive started
📈 1 - Ensemble with avg smape 120.02:
2 - Ensemble with avg smape 120.03:
📈 3 - Ensemble with avg smape 8.19:
4 - Ensemble with avg smape 75.0:
5 - Ensemble with avg smape 19.47:
6 - Ensemble with avg smape 8.92:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Ensemble Horizontal component 1 of 3 ConstantNaive started
Ensemble Horizontal component 2 of 3 ConstantNaive started
Ensemble Horizontal component 3 of 3 LastValueNaive started
[12]:
| PC1 | PC2 | PC3 | PC4 | PC5 | PC6 | PC7 | PC8 | PC9 | PC10 | |
|---|---|---|---|---|---|---|---|---|---|---|
| Date | ||||||||||
| 2023-06-03 | 5.901254 | -2.237191 | 0.598490 | -0.134111 | 0.953554 | 0.01141 | -0.367141 | -0.62687 | 0.080117 | -0.32227 |
| 2023-06-04 | 5.911029 | -2.237191 | 0.600344 | -0.134111 | 0.956330 | 0.01141 | -0.367141 | -0.62687 | 0.081849 | -0.32227 |
| 2023-06-05 | 5.920803 | -2.237191 | 0.602198 | -0.134111 | 0.959105 | 0.01141 | -0.367141 | -0.62687 | 0.083580 | -0.32227 |
| 2023-06-06 | 5.930578 | -2.237191 | 0.604052 | -0.134111 | 0.961881 | 0.01141 | -0.367141 | -0.62687 | 0.085312 | -0.32227 |
| 2023-06-07 | 5.940352 | -2.237191 | 0.605906 | -0.134111 | 0.964656 | 0.01141 | -0.367141 | -0.62687 | 0.087044 | -0.32227 |
2-3. Prediction of ODE parameter values
Now we have Y (estimated ODE parameter values) and X (estimated/forecasted indicator values without multicollinearity), we can predict ODE parameter values of future phases using ODEScenario().predict(days=<int>, name=<str>, seed=0, X=<pandas.DataFrame>). The new scenario is named “Predicted_with_X” here.
[13]:
snr.build_with_template(name="Predicted_with_X", template="Baseline")
snr.predict(days=30, name="Predicted_with_X", seed=0, X=future_df);
Using 1 cpus for n_jobs.
Data frequency is: D, used frequency is: D
Model Number: 1 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "fake_date", "transformations": {"0": "DifferencedTransformer", "1": "SinTrend"}, "transformation_params": {"0": {}, "1": {}}}
Model Number: 2 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 3 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean"} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "Round", "2": "Detrend"}, "transformation_params": {"0": {"lag_1": 7, "method": "Mean"}, "1": {"model": "middle", "decimals": 2, "on_transform": true, "on_inverse": false}, "2": {"model": "GLS"}}}
Model Number: 4 with model GLS in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "RollingMeanTransformer", "1": "DifferencedTransformer", "2": "Detrend", "3": "Slice"}, "transformation_params": {"0": {"fixed": true, "window": 3}, "1": {}, "2": {"model": "Linear"}, "3": {"method": 100}}}
Model Number: 5 with model GLS in generation 0 of 1 with params {} and transformations {"fillna": "median", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "RobustScaler", "3": "Round", "4": "MaxAbsScaler"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}, "3": {"model": "middle", "decimals": 2, "on_transform": true, "on_inverse": true}, "4": {}}}
Model Number: 6 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "bkfilter", "1": "SinTrend", "2": "Detrend", "3": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {}, "2": {"model": "Linear"}, "3": {}}}
Model Number: 7 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PositiveShift", "1": "SinTrend", "2": "bkfilter"}, "transformation_params": {"0": {}, "1": {}, "2": {}}}
Model Number: 8 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "SinTrend"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}}}
Model Number: 9 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 10 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 2, "lag_2": 7} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SinTrend", "1": "Round", "2": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {"model": "middle", "decimals": 2, "on_transform": false, "on_inverse": true}, "2": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6580, in _fit_one
df = self.transformers[i].fit_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/utils/_set_output.py", line 319, in wrapped
data_to_wrap = f(self, X, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/base.py", line 1389, in wrapper
return fit_method(estimator, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py", line 3320, in fit_transform
return self._fit(X, y, force_transform=True)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py", line 3352, in _fit
self.lambdas_[i] = optim_function(col)
^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py", line 3530, in _yeo_johnson_optimize
return optimize.brent(_neg_log_likelihood, brack=(-2, 2))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2626, in brent
res = _minimize_scalar_brent(func, brack, args, **options)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2663, in _minimize_scalar_brent
brent.optimize()
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2433, in optimize
xa, xb, xc, fa, fb, fc, funcalls = self.get_bracket_info()
^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 2402, in get_bracket_info
xa, xb, xc, fa, fb, fc, funcalls = bracket(func, xa=brack[0],
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/scipy/optimize/_optimize.py", line 3032, in bracket
raise e
scipy.optimize._optimize.BracketError: The algorithm terminated without finding a valid bracket. Consider trying different initial points.
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 922, in fit
df_train_transformed = self.transformer_object._fit(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer PowerTransformer failed on fit from params rolling_mean_24 {'0': {}, '1': {'model': 'middle', 'decimals': 2, 'on_transform': False, 'on_inverse': True}, '2': {}} with error BracketError('The algorithm terminated without finding a valid bracket. Consider trying different initial points.')
in model 10 in generation 0: SeasonalNaive
Model Number: 11 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 2, "lag_2": 1} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "SeasonalDifference", "1": "QuantileTransformer", "2": "Detrend"}, "transformation_params": {"0": {"lag_1": 12, "method": "Median"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"model": "GLS"}}}
Model Number: 12 with model SeasonalNaive in generation 0 of 1 with params {"method": "LastValue", "lag_1": 7, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "QuantileTransformer", "1": "ClipOutliers"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 13 with model ConstantNaive in generation 0 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer", "2": "SeasonalDifference"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 7, "method": "LastValue"}}}
/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/preprocessing/_data.py:3475: RuntimeWarning: overflow encountered in power
out[pos] = (np.power(x[pos] + 1, lmbda) - 1) / lmbda
Model Number: 14 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 28} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 15 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "sokalmichener", "include_differenced": true, "k": 20, "stride_size": 1, "regression_type": null} and transformations {"fillna": "zero", "transformations": {"0": null}, "transformation_params": {"0": {}}}
Model Number: 16 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "QuantileTransformer", "1": "QuantileTransformer", "2": "RobustScaler"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 17 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 30} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
Model Number: 18 with model SeasonalityMotif in generation 0 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "mae", "k": 10, "datepart_method": "common_fourier"} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false}}}
Model Number: 19 with model SectionalMotif in generation 0 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "AlignLastValue", "2": "SeasonalDifference", "3": "LevelShiftTransformer"}, "transformation_params": {"0": {"lag_1": 7, "method": "Median"}, "1": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 30, "alpha": 2.0, "grouping_forward_limit": 4, "max_level_shifts": 10, "alignment": "average"}}}
Model Number: 20 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "simple_binarized", "changepoint_spacing": 360, "changepoint_distance_end": 360, "regression_type": null, "lambda_": 0.01} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "Round", "1": "AlignLastValue", "2": "HistoricValues", "3": "ClipOutliers", "4": "bkfilter"}, "transformation_params": {"0": {"decimals": -1, "on_transform": true, "on_inverse": true}, "1": {"rows": 7, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": true, "threshold": null, "threshold_method": "mean"}, "2": {"window": 10}, "3": {"method": "clip", "std_threshold": 3, "fillna": null}, "4": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3620, in fit
noise = np.random.normal(0, 0.001, size=Y_values)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "numpy/random/mtrand.pyx", line 1580, in numpy.random.mtrand.RandomState.normal
File "_common.pyx", line 657, in numpy.random._common.cont
ValueError: maximum supported dimension for an ndarray is currently 64, found 1171
in model 20 in generation 0: BasicLinearModel
Model Number: 21 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["dayofweek", [365.25, 14]], "changepoint_spacing": 90, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98} and transformations {"fillna": "piecewise_polynomial", "transformations": {"0": "AlignLastValue", "1": "IntermittentOccurrence", "2": "RobustScaler", "3": "Log"}, "transformation_params": {"0": {"rows": 1, "lag": 2, "method": "multiplicative", "strength": 0.9, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"center": "mean"}, "2": {}, "3": {}}}
Model Number: 22 with model SeasonalityMotif in generation 0 of 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
Model Number: 23 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "recurring", "changepoint_spacing": 28, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 0.01, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Slice", "1": "ClipOutliers", "2": "SeasonalDifference", "3": "LevelShiftTransformer", "4": "LevelShiftTransformer", "5": "PositiveShift"}, "transformation_params": {"0": {"method": 0.2}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}, "2": {"lag_1": 12, "method": 5}, "3": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "4": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "5": {}}}
Model Number: 24 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "zero", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AnomalyRemoval"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": 30, "transform_dict": null}, "2": {"method": "zscore", "method_params": {"distribution": "uniform", "alpha": 0.14}, "fillna": "fake_date", "transform_dict": {"fillna": "quadratic", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 7, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}}}, "isolated_only": false, "on_inverse": false}}}
Model Number: 25 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastDiff", "1": "MaxAbsScaler", "2": "HolidayTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"rows": 28, "displacement_rows": 1, "quantile": 0.7, "decay_span": null}, "1": {}, "2": {"threshold": 0.8, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": false, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": true, "use_hindu_holidays": false, "anomaly_detector_params": {"method": "med_diff", "method_params": {"distribution": "beta", "alpha": 0.05}, "fillna": "ffill", "transform_dict": null, "isolated_only": true, "on_inverse": false}, "remove_excess_anomalies": false, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 1000}}, "datepart_method": [168, "hour"], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {"rows": 1, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 26 with model AverageValueNaive in generation 0 of 1 with params {"method": "Midhinge", "window": null} and transformations {"fillna": "median", "transformations": {"0": "ScipyFilter", "1": "HPFilter", "2": "AlignLastValue", "3": "UpscaleDownscaleTransformer"}, "transformation_params": {"0": {"method": "savgol_filter", "method_args": {"window_length": 31, "polyorder": 1, "deriv": 0, "mode": "mirror"}}, "1": {"part": "trend", "lamb": 129600}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "3": {"mode": "upscale", "factor": 1, "down_method": "decimate", "fill_method": "linear"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2288, in TemplateWizard
template_result, best_smape = _eval_prediction_for_template(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1637, in _eval_prediction_for_template
model_error = df_forecast.evaluate(
^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/base.py", line 830, in evaluate
) = full_metric_evaluation(
^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/metrics.py", line 630, in full_metric_evaluation
full_errors = F - A
~~^~~
ValueError: operands could not be broadcast together with shapes (60,4) (30,4)
in model 26 in generation 0: AverageValueNaive
Model Number: 27 with model GLS in generation 0 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "ChangepointDetrend"}, "transformation_params": {"0": {"model": "Linear", "changepoint_spacing": 5040, "changepoint_distance_end": 28, "datepart_method": null}}}
Model Number: 28 with model SeasonalNaive in generation 0 of 1 with params {"method": "mean", "lag_1": 2, "lag_2": 60} and transformations {"fillna": "rolling_mean", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}}}
Model Number: 29 with model SeasonalityMotif in generation 0 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 30 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "yule", "include_differenced": true, "k": 3, "stride_size": 1, "regression_type": null, "comparison_transformation": {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "EWMAFilter"}, "transformation_params": {"0": {"rows": null, "displacement_rows": 4, "quantile": 1.0, "decay_span": null}, "1": {"span": 12}}}, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "quadratic", "transformations": {"0": "AlignLastValue", "1": "PCA", "2": "QuantileTransformer", "3": "AlignLastValue", "4": "HPFilter", "5": "CenterSplit"}, "transformation_params": {"0": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "1": {"whiten": false, "n_components": 0.3}, "2": {"output_distribution": "uniform", "n_quantiles": 1000}, "3": {"rows": 168, "lag": 84, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "max"}, "4": {"part": "trend", "lamb": 1600}, "5": {"fillna": "ffill", "center": "median"}}}
Model Number: 31 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 28, "changepoint_distance_end": 180, "regression_type": null, "lambda_": null, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill", "transformations": {"0": "LevelShiftTransformer", "1": "cffilter", "2": "MaxAbsScaler", "3": "SinTrend"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}, "1": {}, "2": {}, "3": {}}}
Model Number: 32 with model SeasonalityMotif in generation 0 of 1 with params {"window": 30, "point_method": "mean", "distance_metric": "mae", "k": 5, "datepart_method": "recurring", "independent": false} and transformations {"fillna": "median", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "RegressionFilter", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 5, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "2": {"sigma": 1, "rolling_window": 90, "run_order": "season_first", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 3, "min_samples_split": 1.0}}, "datepart_method": "expanded_binarized", "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "holiday_params": null, "trend_method": "rolling_mean"}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 33 with model AverageValueNaive in generation 0 of 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 9, "min_samples_split": 1.0}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
Model Number: 34 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 35 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 90, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastDiff", "1": "IntermittentOccurrence", "2": "Discretize", "3": "Discretize", "4": "AlignLastValue", "5": "RollingMeanTransformer"}, "transformation_params": {"0": {"rows": 28, "displacement_rows": 1, "quantile": 1.0, "decay_span": 365}, "1": {"center": "mean"}, "2": {"discretization": "upper", "n_bins": 10}, "3": {"discretization": "center", "n_bins": 20}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.9, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "5": {"fixed": true, "window": 3, "macro_micro": false, "center": true}}}
Model Number: 36 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 37 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "median", "distance_metric": "jaccard", "include_differenced": false, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "rolling_mean", "transformations": {"0": "QuantileTransformer", "1": "AlignLastValue", "2": "StandardScaler", "3": "CumSumTransformer"}, "transformation_params": {"0": {"output_distribution": "normal", "n_quantiles": "fifth"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.7, "first_value_only": false, "threshold": 3, "threshold_method": "max"}, "2": {}, "3": {}}}
Model Number: 38 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 90, "changepoint_distance_end": null, "regression_type": null, "lambda_": 0.01, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "IntermittentOccurrence", "1": "AlignLastValue", "2": "AlignLastValue", "3": "RollingMeanTransformer", "4": "AlignLastValue", "5": "AlignLastValue"}, "transformation_params": {"0": {"center": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "3": {"fixed": false, "window": 12, "macro_micro": true, "center": false}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "5": {"rows": 24, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 39 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {}}}
Model Number: 40 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 41 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "akima", "transformations": {"0": "MinMaxScaler", "1": "SeasonalDifference", "2": "MinMaxScaler", "3": "EWMAFilter", "4": "LevelShiftTransformer", "5": "Detrend"}, "transformation_params": {"0": {}, "1": {"lag_1": 12, "method": "LastValue"}, "2": {}, "3": {"span": 10}, "4": {"window_size": 14, "alpha": 4.0, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "rolling_diff_5nn"}, "5": {"model": "Linear", "phi": 1, "window": null, "transform_dict": null}}}
Model Number: 42 with model SeasonalNaive in generation 0 of 1 with params {"method": "lastvalue", "lag_1": 70, "lag_2": null} and transformations {"fillna": "pchip", "transformations": {"0": "CenterLastValue", "1": "DifferencedTransformer", "2": "Detrend"}, "transformation_params": {"0": {"rows": 3}, "1": {"lag": 1, "fill": "bfill"}, "2": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}}}
Model Number: 43 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 180, "constant": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "SeasonalDifference", "1": "DifferencedTransformer", "2": "AlignLastValue", "3": "AlignLastValue", "4": "Round", "5": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"decimals": 0, "on_transform": false, "on_inverse": true}, "5": {"lag_1": 7, "method": "Median"}}}
Model Number: 44 with model GLS in generation 0 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}}}
Model Number: 45 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "quadratic", "transformations": {"0": "PctChangeTransformer", "1": "MeanPercentSplitter", "2": "StandardScaler"}, "transformation_params": {"0": {}, "1": {"window": 3}, "2": {}}}
Model Number: 46 with model ConstantNaive in generation 0 of 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 47 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
Model Number: 48 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 5040, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 0.001, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}}}
Model Number: 49 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 180, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "RobustScaler", "1": "Discretize", "2": "CenterSplit"}, "transformation_params": {"0": {}, "1": {"discretization": "center", "n_bins": 5}, "2": {"fillna": "mean", "center": "zero"}}}
Model Number: 50 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["day", 168, 52], "changepoint_spacing": 90, "changepoint_distance_end": 90, "regression_type": "User", "lambda_": null, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"lag": 1, "fill": "bfill"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.5, "first_value_only": false, "threshold": 10, "threshold_method": "max"}}}
Model Number: 51 with model SectionalMotif in generation 0 of 1 with params {"window": 5, "point_method": "weighted_mean", "distance_metric": "minkowski", "include_differenced": true, "k": 100, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue", "4": "AlignLastValue", "5": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"model": "Linear", "phi": 0.999, "window": 90, "transform_dict": {"fillna": "rolling_mean_24", "transformations": {"0": "RollingMean100thN", "1": "DifferencedTransformer"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "bfill"}}}}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "5": {"rows": 2}}}
Model Number: 52 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "cityblock", "include_differenced": true, "k": 15, "stride_size": 1, "regression_type": "User", "comparison_transformation": {"fillna": "fake_date", "transformations": {"0": "Detrend", "1": "DifferencedTransformer"}, "transformation_params": {"0": {"model": "GLS", "phi": 0.999, "window": null, "transform_dict": null}, "1": {"lag": 1, "fill": "bfill"}}}, "combination_transformation": null} and transformations {"fillna": "time", "transformations": {"0": "ClipOutliers", "1": "Detrend"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": {"fillna": null, "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4}}}}}}
Model Number: 53 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "expanded_binarized", "changepoint_spacing": 180, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 0.1, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "DatepartRegression", "1": "AlignLastValue", "2": "SinTrend", "3": "MinMaxScaler", "4": "AlignLastValue"}, "transformation_params": {"0": {"regression_model": {"model": "ElasticNet", "model_params": {"l1_ratio": 0.5, "fit_intercept": true, "selection": "cyclic", "max_iter": 5000}}, "datepart_method": [7, 365.25], "polynomial_degree": null, "transform_dict": null, "holiday_countries_used": true, "lags": null, "forward_lags": null}, "1": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {"rows": 1, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 930, in fit
self.model = self.model.fit(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/models/basics.py", line 3627, in fit
np.linalg.inv(X_values.T @ X_values + self.lambda_ * I)
ValueError: matmul: Input operand 1 has a mismatch in its core dimension 0, with gufunc signature (n?,k),(k,m?)->(n?,m?) (size 1171 is different from 2436)
in model 53 in generation 0: BasicLinearModel
Model Number: 54 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": null, "constant": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "STLFilter", "1": "MinMaxScaler", "2": "StandardScaler"}, "transformation_params": {"0": {"decomp_type": "STL", "part": "trend", "seasonal": 7}, "1": {}, "2": {}}}
Model Number: 55 with model ConstantNaive in generation 0 of 1 with params {"constant": 1} and transformations {"fillna": "pchip", "transformations": {"0": "AlignLastValue", "1": "ThetaTransformer", "2": "DifferencedTransformer", "3": "StandardScaler", "4": "DifferencedTransformer"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "max"}, "1": {"theta_values": [0.8, 1.2]}, "2": {"lag": 1, "fill": "bfill"}, "3": {}, "4": {"lag": 1, "fill": "zero"}}}
Model Number: 56 with model LastValueNaive in generation 0 of 1 with params {} and transformations {"fillna": "pchip", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "DifferencedTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 4, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 365, "transform_dict": null}, "2": {"lag": 1, "fill": "bfill"}, "3": {"rows": 168, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 57 with model AverageValueNaive in generation 0 of 1 with params {"method": "Median", "window": null} and transformations {"fillna": "ffill", "transformations": {"0": "QuantileTransformer"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": "quarter"}}}
Model Number: 58 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["morlet_365.25_12_12", "ricker_7_7_1"], "changepoint_spacing": 60, "changepoint_distance_end": 360, "regression_type": null, "lambda_": null, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "Detrend", "1": "IntermittentOccurrence", "2": "MeanPercentSplitter"}, "transformation_params": {"0": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "1": {"center": "mean"}, "2": {"window": "forecast_length"}}}
Model Number: 59 with model SectionalMotif in generation 0 of 1 with params {"window": 10, "point_method": "midhinge", "distance_metric": "correlation", "include_differenced": true, "k": 1, "stride_size": 1, "regression_type": "User", "comparison_transformation": {"fillna": "quadratic", "transformations": {"0": "Log"}, "transformation_params": {"0": {}}}, "combination_transformation": null} and transformations {"fillna": "time", "transformations": {"0": "FFTFilter"}, "transformation_params": {"0": {"cutoff": 0.6, "reverse": false, "on_transform": true, "on_inverse": false}}}
Model Number: 60 with model GLS in generation 0 of 1 with params {"changepoint_spacing": 5040, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "quadratic", "transformations": {"0": "ClipOutliers"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 2, "fillna": null}}}
Model Number: 61 with model AverageValueNaive in generation 0 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "zero", "transformations": {"0": "AnomalyRemoval", "1": "Constraint", "2": "MinMaxScaler"}, "transformation_params": {"0": {"method": "minmax", "method_params": {"alpha": 0.03}, "fillna": "ffill", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HistoricValues", "1": "PowerTransformer"}, "transformation_params": {"0": {"window": null}, "1": {}}}, "isolated_only": false, "on_inverse": false}, "1": {"constraint_method": "stdev_min", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 3.0, "bounds_only": false, "fillna": "ffill"}, "2": {}}}
Model Number: 62 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": ["weekdayofmonth", "common_fourier"], "changepoint_spacing": 180, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 100, "trend_phi": null, "holiday_countries_used": true} and transformations {"fillna": "pchip", "transformations": {"0": "STLFilter", "1": "AlignLastValue"}, "transformation_params": {"0": {"decomp_type": "seasonal_decompose", "part": "trend"}, "1": {"rows": 84, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 63 with model BasicLinearModel in generation 0 of 1 with params {"datepart_method": "common_fourier", "changepoint_spacing": 60, "changepoint_distance_end": 520, "regression_type": null, "lambda_": 1, "trend_phi": 0.97, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}}}
New Generation: 1 of 1
Model Number: 64 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "SeasonalDifference", "1": "AlignLastValue", "2": "AnomalyRemoval", "3": "AnomalyRemoval"}, "transformation_params": {"0": {"lag_1": 91, "method": "LastValue"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "2": {"method": "rolling_zscore", "method_params": {"distribution": "uniform", "alpha": 0.05, "rolling_periods": 28, "center": true}, "fillna": "ffill", "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}, "isolated_only": false, "on_inverse": true}, "3": {"method": "rolling_zscore", "method_params": {"distribution": "chi2", "alpha": 0.1, "rolling_periods": 90, "center": true}, "fillna": "linear", "transform_dict": {"transformations": {"0": "DifferencedTransformer"}, "transformation_params": {"0": {}}}, "isolated_only": false, "on_inverse": false}}}
Model Number: 65 with model SeasonalityMotif in generation 1 of 1 with params {"window": 50, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "mean", "transformations": {"0": "Constraint", "1": "AlignLastValue", "2": "SeasonalDifference"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"rows": 168, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {"lag_1": 364, "method": "Mean"}}}
Model Number: 66 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "closest", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": false} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "SeasonalDifference"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"lag_1": 364, "method": "Mean"}}}
Model Number: 67 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": null} and transformations {"fillna": "mean", "transformations": {"0": "Discretize"}, "transformation_params": {"0": {"discretization": "lower", "n_bins": 50}}}
Model Number: 68 with model SeasonalNaive in generation 1 of 1 with params {"method": "mean", "lag_1": 70, "lag_2": null} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "RobustScaler"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {}}}
Model Number: 69 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 70 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "ShiftFirstValue", "1": "AlignLastValue", "2": "MeanPercentSplitter", "3": "FFTFilter", "4": "SeasonalDifference"}, "transformation_params": {"0": {"rows": 2}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}, "2": {"window": 7}, "3": {"cutoff": 0.05, "reverse": false, "on_transform": true, "on_inverse": true}, "4": {"lag_1": 28, "method": 20}}}
Model Number: 71 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": "recurring", "changepoint_spacing": 28, "changepoint_distance_end": 90, "regression_type": null, "lambda_": 100, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "AlignLastValue", "1": "SeasonalDifference", "2": "RobustScaler", "3": "Round"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"lag_1": 2, "method": "Mean"}, "2": {}, "3": {"decimals": 2, "on_transform": false, "on_inverse": true}}}
Model Number: 72 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "MinMaxScaler", "1": "AlignLastValue", "2": "LevelShiftTransformer", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {}, "1": {"rows": 168, "lag": 1, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"window_size": 364, "alpha": 3.0, "grouping_forward_limit": 3, "max_level_shifts": 10, "alignment": "average"}, "3": {}, "4": {}, "5": {}}}
Model Number: 73 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": ["dayofmonthofyear", "week", 7], "changepoint_spacing": 60, "changepoint_distance_end": 60, "regression_type": null, "lambda_": 1, "trend_phi": null, "holiday_countries_used": false} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "LevelShiftTransformer", "2": "PositiveShift"}, "transformation_params": {"0": {"lag_1": 12, "method": "LastValue"}, "1": {"window_size": 7, "alpha": 2.0, "grouping_forward_limit": 5, "max_level_shifts": 30, "alignment": "rolling_diff_3nn"}, "2": {}}}
Model Number: 74 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "pchip", "transformations": {"0": "MinMaxScaler", "1": "PctChangeTransformer", "2": "AlignLastValue", "3": "PositiveShift"}, "transformation_params": {"0": {}, "1": {}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {}}}
Model Number: 75 with model GLS in generation 1 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "DifferencedTransformer", "2": "Slice"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {}, "2": {"method": 100}}}
Model Number: 76 with model SectionalMotif in generation 1 of 1 with params {"window": 5, "point_method": "median", "distance_metric": "minkowski", "include_differenced": true, "k": 100, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "rolling_mean", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 77 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "ffill_mean_biased", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 3, "threshold_method": "mean"}}}
Model Number: 78 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "mse", "k": 20, "datepart_method": "lunar_phase", "independent": true} and transformations {"fillna": "nearest", "transformations": {"0": "AnomalyRemoval"}, "transformation_params": {"0": {"method": "mad", "method_params": {"distribution": "uniform", "alpha": 0.05}, "fillna": "linear", "transform_dict": {"transformations": {"0": "DatepartRegression"}, "transformation_params": {"0": {"datepart_method": "simple_3", "regression_model": {"model": "DecisionTree", "model_params": {"max_depth": null, "min_samples_split": 0.1}}}}}, "isolated_only": true, "on_inverse": false}}}
Model Number: 79 with model SeasonalityMotif in generation 1 of 1 with params {"window": 5, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "AlignLastValue", "3": "ScipyFilter", "4": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": null, "threshold_method": "max"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 1, "threshold_method": "mean"}, "3": {"method": "butter", "method_args": {"N": 4, "window_size": 7, "btype": "lowpass", "analog": false, "output": "sos"}}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
Model Number: 80 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "median", "distance_metric": "canberra", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 81 with model ConstantNaive in generation 1 of 1 with params {"constant": 1} and transformations {"fillna": "fake_date", "transformations": {"0": "ClipOutliers", "1": "Detrend"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}}}
Model Number: 82 with model ConstantNaive in generation 1 of 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "EWMAFilter", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {"span": 12}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 83 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "ffill", "transformations": {"0": "MinMaxScaler", "1": "AlignLastValue", "2": "bkfilter", "3": "PositiveShift", "4": "StandardScaler", "5": "SeasonalDifference"}, "transformation_params": {"0": {}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}, "2": {}, "3": {}, "4": {}, "5": {"lag_1": 2, "method": "LastValue"}}}
Model Number: 84 with model SectionalMotif in generation 1 of 1 with params {"window": 7, "point_method": "weighted_mean", "distance_metric": "mahalanobis", "include_differenced": true, "k": 5, "stride_size": 1, "regression_type": "common_fourier", "comparison_transformation": null, "combination_transformation": null} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "CenterLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "2": {"rows": 2}}}
Model Number: 85 with model SeasonalityMotif in generation 1 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 5, "datepart_method": "simple", "independent": true} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue", "2": "ScipyFilter", "3": "AlignLastValue", "4": "HPFilter"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": null, "threshold_method": "max"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 1, "threshold_method": "mean"}, "2": {"method": "butter", "method_args": {"N": 4, "window_size": 7, "btype": "lowpass", "analog": false, "output": "sos"}}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "4": {"part": "trend", "lamb": 1600}}}
Model Number: 86 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 70, "lag_2": null} and transformations {"fillna": "nearest", "transformations": {"0": "QuantileTransformer", "1": "Cointegration", "2": "FFTDecomposition"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"det_order": 1, "k_ar_diff": 1}, "2": {"n_harmonics": 10, "detrend": null}}}
Model Number: 87 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
Model Number: 88 with model GLS in generation 1 of 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "AlignLastValue", "1": "STLFilter"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"decomp_type": "STL", "part": "resid", "seasonal": 167}}}
Model Number: 89 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "DifferencedTransformer", "3": "AlignLastValue", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": 10}, "1": {"rows": 1, "lag": 84, "method": "additive", "strength": 0.7, "first_value_only": true, "threshold": null, "threshold_method": "mean"}, "2": {}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.7, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "mean"}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6580, in _fit_one
df = self.transformers[i].fit_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 2588, in fit_transform
return self._fit(df)
^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 2528, in _fit
return_df = self.transformer.fit_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/utils/_set_output.py", line 319, in wrapped
data_to_wrap = f(self, X, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/base.py", line 1389, in wrapper
return fit_method(estimator, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/decomposition/_pca.py", line 468, in fit_transform
U, S, _, X, x_is_centered, xp = self._fit(X)
^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/decomposition/_pca.py", line 542, in _fit
return self._fit_full(X, n_components, xp, is_array_api_compliant)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/decomposition/_pca.py", line 556, in _fit_full
raise ValueError(
ValueError: n_components=10 must be between 0 and min(n_samples, n_features)=4 with svd_solver='covariance_eigh'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 922, in fit
df_train_transformed = self.transformer_object._fit(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer PCA failed on fit from params fake_date {'0': {'whiten': False, 'n_components': 10}, '1': {'rows': 1, 'lag': 84, 'method': 'additive', 'strength': 0.7, 'first_value_only': True, 'threshold': None, 'threshold_method': 'mean'}, '2': {}, '3': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 0.7, 'first_value_only': False, 'threshold': None, 'threshold_method': 'max'}, '4': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}} with error ValueError("n_components=10 must be between 0 and min(n_samples, n_features)=4 with svd_solver='covariance_eigh'")
in model 89 in generation 1: AverageValueNaive
Model Number: 90 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "bkfilter", "1": "Detrend", "2": "PowerTransformer"}, "transformation_params": {"0": {}, "1": {"model": "Linear"}, "2": {}}}
Model Number: 91 with model SectionalMotif in generation 1 of 1 with params {"window": 10, "point_method": "median", "distance_metric": "hamming", "include_differenced": true, "k": 10, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "ffill", "transformations": {"0": "QuantileTransformer", "1": "RobustScaler"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {}}}
Model Number: 92 with model SeasonalityMotif in generation 1 of 1 with params {"window": 7, "point_method": "midhinge", "distance_metric": "chebyshev", "k": 10, "datepart_method": "common_fourier_rw", "independent": false} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "UpscaleDownscaleTransformer", "1": "Detrend", "2": "AnomalyRemoval", "3": "AlignLastValue", "4": "AlignLastValue", "5": "ShiftFirstValue"}, "transformation_params": {"0": {"mode": "downscale", "factor": 2, "down_method": "decimate", "fill_method": "linear"}, "1": {"model": "GLS", "phi": 1, "window": null, "transform_dict": null}, "2": {"method": "rolling_zscore", "method_params": {"distribution": "uniform", "alpha": 0.03, "rolling_periods": 90, "center": true}, "fillna": "rolling_mean_24", "transform_dict": null, "isolated_only": false, "on_inverse": false}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "5": {"rows": 28}}}
2023-05-09 00:00:00
2023-05-09 00:00:00
2023-05-09 00:00:00
Model Number: 93 with model GLS in generation 1 of 1 with params {"changepoint_spacing": 180, "changepoint_distance_end": 90, "constant": false} and transformations {"fillna": "zero", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {}, "2": {}}}
Model Number: 94 with model GLS in generation 1 of 1 with params {"changepoint_spacing": 60, "changepoint_distance_end": 28, "constant": true} and transformations {"fillna": "ffill", "transformations": {"0": "RobustScaler", "1": "LevelShiftTransformer"}, "transformation_params": {"0": {}, "1": {"window_size": 90, "alpha": 4.0, "grouping_forward_limit": 6, "max_level_shifts": 5, "alignment": "rolling_diff_3nn"}}}
Model Number: 95 with model SeasonalNaive in generation 1 of 1 with params {"method": "lastvalue", "lag_1": 7, "lag_2": 60} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "QuantileTransformer"}, "transformation_params": {"0": {"lag_1": 12, "method": "Median"}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
Model Number: 96 with model SeasonalityMotif in generation 1 of 1 with params {"window": 7, "point_method": "trimmed_mean_40", "distance_metric": "chebyshev", "k": 5, "datepart_method": "simple", "independent": true} and transformations {"fillna": "ffill", "transformations": {"0": "ThetaTransformer", "1": "AlignLastDiff", "2": "MaxAbsScaler", "3": "AlignLastValue", "4": "ShiftFirstValue"}, "transformation_params": {"0": {"theta_values": [0.5, 1.5]}, "1": {"rows": 90, "displacement_rows": 1, "quantile": 1.0, "decay_span": 4}, "2": {}, "3": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "4": {"rows": 28}}}
Model Number: 97 with model SectionalMotif in generation 1 of 1 with params {"window": 10, "point_method": "weighted_mean", "distance_metric": "canberra", "include_differenced": true, "k": 3, "stride_size": 1, "regression_type": null, "comparison_transformation": null, "combination_transformation": {"fillna": "ffill", "transformations": {"0": "StandardScaler"}, "transformation_params": {"0": {}}}} and transformations {"fillna": "nearest", "transformations": {"0": "AlignLastValue", "1": "FIRFilter", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "1": {"numtaps": 7, "cutoff_hz": 0.01, "window": "kaiser", "sampling_frequency": 12, "on_transform": true, "on_inverse": false, "bounds_only": false}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}, "3": {"rows": 84, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 10, "threshold_method": "max"}}}
Model Number: 98 with model ConstantNaive in generation 1 of 1 with params {"constant": 1} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "Detrend", "2": "AlignLastValue", "3": "AlignLastValue"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3.5, "fillna": null}, "1": {"model": "Linear", "phi": 1, "window": 30, "transform_dict": {"fillna": null, "transformations": {"0": "EWMAFilter"}, "transformation_params": {"0": {"span": 7}}}}, "2": {"rows": 4, "lag": 2, "method": "multiplicative", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "3": {"rows": 1, "lag": 7, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}}}
Model Number: 99 with model SeasonalityMotif in generation 1 of 1 with params {"window": 5, "point_method": "midhinge", "distance_metric": "canberra", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
Model Number: 100 with model BasicLinearModel in generation 1 of 1 with params {"datepart_method": ["weekdaymonthofyear", "quarter", "dayofweek"], "changepoint_spacing": 6, "changepoint_distance_end": 360, "regression_type": null, "lambda_": 10000, "trend_phi": 0.98, "holiday_countries_used": true} and transformations {"fillna": "ffill", "transformations": {"0": "LevelShiftTransformer", "1": "IntermittentOccurrence", "2": "MaxAbsScaler", "3": "SinTrend"}, "transformation_params": {"0": {"window_size": 30, "alpha": 3.5, "grouping_forward_limit": 4, "max_level_shifts": 5, "alignment": "last_value"}, "1": {"center": "mean"}, "2": {}, "3": {}}}
Model Number: 101 with model AverageValueNaive in generation 1 of 1 with params {"method": "Weighted_Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
Model Number: 102 with model AverageValueNaive in generation 1 of 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "Discretize", "3": "cffilter"}, "transformation_params": {"0": {"whiten": false, "n_components": 10}, "1": {"rows": 1, "lag": 84, "method": "additive", "strength": 0.7, "first_value_only": true, "threshold": null, "threshold_method": "mean"}, "2": {"discretization": "lower", "n_bins": 50}, "3": {}}}
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6580, in _fit_one
df = self.transformers[i].fit_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 2588, in fit_transform
return self._fit(df)
^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 2528, in _fit
return_df = self.transformer.fit_transform(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/utils/_set_output.py", line 319, in wrapped
data_to_wrap = f(self, X, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/base.py", line 1389, in wrapper
return fit_method(estimator, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/decomposition/_pca.py", line 468, in fit_transform
U, S, _, X, x_is_centered, xp = self._fit(X)
^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/decomposition/_pca.py", line 542, in _fit
return self._fit_full(X, n_components, xp, is_array_api_compliant)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/sklearn/decomposition/_pca.py", line 556, in _fit_full
raise ValueError(
ValueError: n_components=10 must be between 0 and min(n_samples, n_features)=4 with svd_solver='covariance_eigh'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1597, in model_forecast
model = model.fit(df_train_low, future_regressor_train)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 922, in fit
df_train_transformed = self.transformer_object._fit(df)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer PCA failed on fit from params fake_date {'0': {'whiten': False, 'n_components': 10}, '1': {'rows': 1, 'lag': 84, 'method': 'additive', 'strength': 0.7, 'first_value_only': True, 'threshold': None, 'threshold_method': 'mean'}, '2': {'discretization': 'lower', 'n_bins': 50}, '3': {}} with error ValueError("n_components=10 must be between 0 and min(n_samples, n_features)=4 with svd_solver='covariance_eigh'")
in model 102 in generation 1: AverageValueNaive
Model Number: 103 with model LastValueNaive in generation 1 of 1 with params {} and transformations {"fillna": "akima", "transformations": {"0": "AlignLastValue", "1": "UpscaleDownscaleTransformer", "2": "CenterSplit", "3": "PositiveShift", "4": "StandardScaler", "5": "bkfilter"}, "transformation_params": {"0": {"rows": 1, "lag": 28, "method": "multiplicative", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}, "1": {"mode": "downscale", "factor": 1, "down_method": "mean", "fill_method": "pchip"}, "2": {"fillna": "ffill", "center": "zero"}, "3": {}, "4": {}, "5": {}}}
2023-05-09 00:00:00
2023-05-09 00:00:00
2023-05-09 00:00:00
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2261, in TemplateWizard
df_forecast = model_forecast(
^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 1598, in model_forecast
return model.predict(
^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 972, in predict
raise ValueError(
ValueError: Model LastValueNaive returned improper forecast_length. Returned: 28 and requested: 30
in model 103 in generation 1: LastValueNaive
Validation Round: 1
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-03-30', '2023-03-31', '2023-04-01', '2023-04-02',
'2023-04-03', '2023-04-04', '2023-04-05', '2023-04-06',
'2023-04-07', '2023-04-08'],
dtype='datetime64[ns]', name='Date', length=1141, freq=None)
Model Number: 1 of 16 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
📈 1 - ConstantNaive with avg smape 24.36:
Model Number: 2 of 16 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
2 - LastValueNaive with avg smape 24.36:
Model Number: 3 of 16 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "pchip", "transformations": {"0": "MinMaxScaler", "1": "PctChangeTransformer", "2": "AlignLastValue", "3": "PositiveShift"}, "transformation_params": {"0": {}, "1": {}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {}}}
3 - LastValueNaive with avg smape 157.8:
Model Number: 4 of 16 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer", "2": "SeasonalDifference"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 7, "method": "LastValue"}}}
4 - ConstantNaive with avg smape 56.87:
Model Number: 5 of 16 with model SeasonalityMotif for Validation 1 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
5 - SeasonalityMotif with avg smape 53.21:
Model Number: 6 of 16 with model SeasonalityMotif for Validation 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
6 - SeasonalityMotif with avg smape 24.36:
Model Number: 7 of 16 with model LastValueNaive for Validation 1 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
7 - LastValueNaive with avg smape 24.36:
Model Number: 8 of 16 with model AverageValueNaive for Validation 1 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 9, "min_samples_split": 1.0}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
8 - AverageValueNaive with avg smape 24.36:
Model Number: 9 of 16 with model GLS for Validation 1 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "AlignLastValue", "1": "STLFilter"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"decomp_type": "STL", "part": "resid", "seasonal": 167}}}
9 - GLS with avg smape 24.36:
Model Number: 10 of 16 with model SeasonalNaive for Validation 1 with params {"method": "lastvalue", "lag_1": 7, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "QuantileTransformer", "1": "ClipOutliers"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"method": "clip", "std_threshold": 2, "fillna": null}}}
10 - SeasonalNaive with avg smape 50.54:
Model Number: 11 of 16 with model SeasonalNaive for Validation 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 28} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
11 - SeasonalNaive with avg smape 70.02:
Model Number: 12 of 16 with model SeasonalNaive for Validation 1 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 30} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
12 - SeasonalNaive with avg smape 68.03:
Model Number: 13 of 16 with model AverageValueNaive for Validation 1 with params {"method": "Weighted_Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
13 - AverageValueNaive with avg smape 24.36:
Model Number: 14 of 16 with model AverageValueNaive for Validation 1 with params {"method": "Mean", "window": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
14 - AverageValueNaive with avg smape 26.28:
Model Number: 15 of 16 with model SeasonalityMotif for Validation 1 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
15 - SeasonalityMotif with avg smape 60.11:
Model Number: 16 of 16 with model ConstantNaive for Validation 1 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
16 - ConstantNaive with avg smape 171.99:
Validation Round: 2
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-02-28', '2023-03-01', '2023-03-02', '2023-03-03',
'2023-03-04', '2023-03-05', '2023-03-06', '2023-03-07',
'2023-03-08', '2023-03-09'],
dtype='datetime64[ns]', name='Date', length=1111, freq=None)
Model Number: 1 of 16 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
📈 1 - ConstantNaive with avg smape 36.13:
Model Number: 2 of 16 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "rolling_mean", "transformations": {"0": "PctChangeTransformer", "1": "RobustScaler", "2": "Round"}, "transformation_params": {"0": {}, "1": {}, "2": {"decimals": -1, "on_transform": true, "on_inverse": false}}}
2 - LastValueNaive with avg smape 36.13:
Model Number: 3 of 16 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "pchip", "transformations": {"0": "MinMaxScaler", "1": "PctChangeTransformer", "2": "AlignLastValue", "3": "PositiveShift"}, "transformation_params": {"0": {}, "1": {}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "max"}, "3": {}}}
3 - LastValueNaive with avg smape 36.13:
Model Number: 4 of 16 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "ffill", "transformations": {"0": "PowerTransformer", "1": "QuantileTransformer", "2": "SeasonalDifference"}, "transformation_params": {"0": {}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {"lag_1": 7, "method": "LastValue"}}}
4 - ConstantNaive with avg smape 36.13:
Model Number: 5 of 16 with model SeasonalityMotif for Validation 2 with params {"window": 5, "point_method": "trimmed_mean_40", "distance_metric": "mae", "k": 5, "datepart_method": ["simple_binarized_poly"], "independent": true} and transformations {"fillna": "rolling_mean", "transformations": {"0": "Constraint", "1": "QuantileTransformer", "2": "AlignLastValue"}, "transformation_params": {"0": {"constraint_method": "historic_diff", "constraint_direction": "lower", "constraint_regularization": 1.0, "constraint_value": 0.2, "bounds_only": false, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 43}, "2": {"rows": 1, "lag": 1, "method": "additive", "strength": 0.2, "first_value_only": false, "threshold": null, "threshold_method": "max"}}}
📈 5 - SeasonalityMotif with avg smape 35.69:
Model Number: 6 of 16 with model SeasonalityMotif for Validation 2 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "ffill", "transformations": {"0": "AlignLastValue", "1": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
📈 6 - SeasonalityMotif with avg smape 33.59:
Model Number: 7 of 16 with model LastValueNaive for Validation 2 with params {} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 1, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}}}
7 - LastValueNaive with avg smape 35.43:
Model Number: 8 of 16 with model AverageValueNaive for Validation 2 with params {"method": "trimmed_mean_40", "window": 28} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue", "1": "PositiveShift", "2": "HolidayTransformer", "3": "PositiveShift", "4": "StandardScaler"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "max"}, "1": {}, "2": {"threshold": 0.9, "splash_threshold": null, "use_dayofmonth_holidays": true, "use_wkdom_holidays": true, "use_wkdeom_holidays": false, "use_lunar_holidays": false, "use_lunar_weekday": false, "use_islamic_holidays": false, "use_hebrew_holidays": false, "use_hindu_holidays": true, "anomaly_detector_params": {"method": "IQR", "method_params": {"iqr_threshold": 2.0, "iqr_quantiles": [0.25, 0.75]}, "fillna": "fake_date", "transform_dict": {"fillna": "ffill", "transformations": {"0": "HPFilter", "1": "ClipOutliers"}, "transformation_params": {"0": {"part": "trend", "lamb": 129600}, "1": {"method": "clip", "std_threshold": 3, "fillna": null}}}, "isolated_only": false, "on_inverse": false}, "remove_excess_anomalies": true, "impact": "datepart_regression", "regression_params": {"regression_model": {"model": "DecisionTree", "model_params": {"max_depth": 9, "min_samples_split": 1.0}}, "datepart_method": "recurring", "polynomial_degree": 2, "transform_dict": {"fillna": null, "transformations": {"0": "bkfilter"}, "transformation_params": {"0": {}}}, "holiday_countries_used": false, "lags": null, "forward_lags": null}}, "3": {}, "4": {}}}
8 - AverageValueNaive with avg smape 36.13:
Model Number: 9 of 16 with model GLS for Validation 2 with params {"changepoint_spacing": null, "changepoint_distance_end": null, "constant": false} and transformations {"fillna": "rolling_mean", "transformations": {"0": "AlignLastValue", "1": "STLFilter"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}, "1": {"decomp_type": "STL", "part": "resid", "seasonal": 167}}}
9 - GLS with avg smape 36.13:
Model Number: 10 of 16 with model SeasonalNaive for Validation 2 with params {"method": "lastvalue", "lag_1": 7, "lag_2": 2} and transformations {"fillna": "mean", "transformations": {"0": "QuantileTransformer", "1": "ClipOutliers"}, "transformation_params": {"0": {"output_distribution": "uniform", "n_quantiles": 1000}, "1": {"method": "clip", "std_threshold": 2, "fillna": null}}}
10 - SeasonalNaive with avg smape 36.13:
Model Number: 11 of 16 with model SeasonalNaive for Validation 2 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 28} and transformations {"fillna": "ffill", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
11 - SeasonalNaive with avg smape 36.13:
Model Number: 12 of 16 with model SeasonalNaive for Validation 2 with params {"method": "lastvalue", "lag_1": 364, "lag_2": 30} and transformations {"fillna": "fake_date", "transformations": {"0": "SeasonalDifference", "1": "MaxAbsScaler", "2": "Round"}, "transformation_params": {"0": {"lag_1": 7, "method": "LastValue"}, "1": {}, "2": {"decimals": 0, "on_transform": false, "on_inverse": true}}}
12 - SeasonalNaive with avg smape 36.13:
Model Number: 13 of 16 with model AverageValueNaive for Validation 2 with params {"method": "Weighted_Mean", "window": null} and transformations {"fillna": "fake_date", "transformations": {"0": "AlignLastValue"}, "transformation_params": {"0": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
13 - AverageValueNaive with avg smape 36.13:
Model Number: 14 of 16 with model AverageValueNaive for Validation 2 with params {"method": "Mean", "window": null} and transformations {"fillna": "mean", "transformations": {"0": "ClipOutliers", "1": "QuantileTransformer", "2": "DifferencedTransformer"}, "transformation_params": {"0": {"method": "clip", "std_threshold": 3, "fillna": null}, "1": {"output_distribution": "uniform", "n_quantiles": 1000}, "2": {}}}
14 - AverageValueNaive with avg smape 40.0:
Model Number: 15 of 16 with model SeasonalityMotif for Validation 2 with params {"window": 10, "point_method": "median", "distance_metric": "mse", "k": 10, "datepart_method": "simple_binarized", "independent": true} and transformations {"fillna": "rolling_mean_24", "transformations": {"0": "PCA", "1": "AlignLastValue", "2": "ThetaTransformer", "3": "QuantileTransformer", "4": "AlignLastValue"}, "transformation_params": {"0": {"whiten": false, "n_components": null}, "1": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": 1, "threshold_method": "mean"}, "2": {"theta_values": [0.6, 1.4]}, "3": {"output_distribution": "uniform", "n_quantiles": 1000}, "4": {"rows": 1, "lag": 1, "method": "additive", "strength": 1.0, "first_value_only": true, "threshold": 10, "threshold_method": "mean"}}}
📈 15 - SeasonalityMotif with avg smape 32.8:
Model Number: 16 of 16 with model ConstantNaive for Validation 2 with params {"constant": 0} and transformations {"fillna": "cubic", "transformations": {"0": "StandardScaler", "1": "DifferencedTransformer", "2": "PctChangeTransformer", "3": "AlignLastValue"}, "transformation_params": {"0": {}, "1": {"lag": 1, "fill": "zero"}, "2": {}, "3": {"rows": 1, "lag": 2, "method": "additive", "strength": 1.0, "first_value_only": false, "threshold": null, "threshold_method": "mean"}}}
16 - ConstantNaive with avg smape 34.06:
Validation Round: 0
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-04-26', '2023-04-27', '2023-04-28', '2023-04-29',
'2023-04-30', '2023-05-01', '2023-05-02', '2023-05-03',
'2023-05-04', '2023-05-05'],
dtype='datetime64[ns]', name='Date', length=1168, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 1 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 3, "model_metric": "Score", "models": {"7a14af550afa2194472cfc2e4e1440fb": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"mean\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"QuantileTransformer\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 1, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 1000}}}"}, "7fb9afb39d6fa9b428cf0f629c21bf27": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 5, \"point_method\": \"trimmed_mean_40\", \"distance_metric\": \"mae\", \"k\": 5, \"datepart_method\": [\"simple_binarized_poly\"], \"independent\": true}", "TransformationParameters": "{\"fillna\": \"rolling_mean\", \"transformations\": {\"0\": \"Constraint\", \"1\": \"QuantileTransformer\", \"2\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"constraint_method\": \"historic_diff\", \"constraint_direction\": \"lower\", \"constraint_regularization\": 1.0, \"constraint_value\": 0.2, \"bounds_only\": false, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 43}, \"2\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 0.2, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"max\"}}}"}, "fd67d39f045cb25329dd5fedf940a41b": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 10, \"point_method\": \"median\", \"distance_metric\": \"mse\", \"k\": 10, \"datepart_method\": \"simple_binarized\", \"independent\": true}", "TransformationParameters": "{\"fillna\": \"ffill\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}, \"1\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}}}"}}, "series": {"theta": "7fb9afb39d6fa9b428cf0f629c21bf27", "kappa": "7a14af550afa2194472cfc2e4e1440fb", "rho": "fd67d39f045cb25329dd5fedf940a41b", "sigma": "7a14af550afa2194472cfc2e4e1440fb"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 3 LastValueNaive started
Ensemble Horizontal component 2 of 3 SeasonalityMotif started
Ensemble Horizontal component 3 of 3 SeasonalityMotif started
📈 1 - Ensemble with avg smape 0.14:
2 - Ensemble with avg smape 0.14:
3 - Ensemble with avg smape 20.98:
4 - Ensemble with avg smape 0.14:
5 - Ensemble with avg smape 10.08:
📈 6 - Ensemble with avg smape 0.0:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Validation Round: 1
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-03-28', '2023-03-29', '2023-03-30', '2023-03-31',
'2023-04-01', '2023-04-02', '2023-04-03', '2023-04-04',
'2023-04-05', '2023-04-06'],
dtype='datetime64[ns]', name='Date', length=1139, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 2 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 3, "model_metric": "Score", "models": {"7a14af550afa2194472cfc2e4e1440fb": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"mean\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"QuantileTransformer\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 1, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 1000}}}"}, "7fb9afb39d6fa9b428cf0f629c21bf27": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 5, \"point_method\": \"trimmed_mean_40\", \"distance_metric\": \"mae\", \"k\": 5, \"datepart_method\": [\"simple_binarized_poly\"], \"independent\": true}", "TransformationParameters": "{\"fillna\": \"rolling_mean\", \"transformations\": {\"0\": \"Constraint\", \"1\": \"QuantileTransformer\", \"2\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"constraint_method\": \"historic_diff\", \"constraint_direction\": \"lower\", \"constraint_regularization\": 1.0, \"constraint_value\": 0.2, \"bounds_only\": false, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 43}, \"2\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 0.2, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"max\"}}}"}, "fd67d39f045cb25329dd5fedf940a41b": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 10, \"point_method\": \"median\", \"distance_metric\": \"mse\", \"k\": 10, \"datepart_method\": \"simple_binarized\", \"independent\": true}", "TransformationParameters": "{\"fillna\": \"ffill\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}, \"1\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}}}"}}, "series": {"theta": "7fb9afb39d6fa9b428cf0f629c21bf27", "kappa": "7a14af550afa2194472cfc2e4e1440fb", "rho": "fd67d39f045cb25329dd5fedf940a41b", "sigma": "7a14af550afa2194472cfc2e4e1440fb"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 3 LastValueNaive started
Ensemble Horizontal component 2 of 3 SeasonalityMotif started
Ensemble Horizontal component 3 of 3 SeasonalityMotif started
📈 1 - Ensemble with avg smape 56.12:
📈 2 - Ensemble with avg smape 56.0:
3 - Ensemble with avg smape 58.97:
4 - Ensemble with avg smape 58.88:
5 - Ensemble with avg smape 59.21:
6 - Ensemble with avg smape 59.31:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Validation Round: 2
Validation train index is DatetimeIndex(['2020-02-23', '2020-02-24', '2020-02-25', '2020-02-26',
'2020-02-27', '2020-02-28', '2020-02-29', '2020-03-01',
'2020-03-02', '2020-03-03',
...
'2023-02-27', '2023-02-28', '2023-03-01', '2023-03-02',
'2023-03-03', '2023-03-04', '2023-03-05', '2023-03-06',
'2023-03-07', '2023-03-08'],
dtype='datetime64[ns]', name='Date', length=1110, freq=None)
Model Number: 1 of 1 with model Ensemble for Validation 3 horizontal ensemble validations with params {"model_name": "Horizontal", "model_count": 3, "model_metric": "Score", "models": {"7a14af550afa2194472cfc2e4e1440fb": {"Model": "LastValueNaive", "ModelParameters": "{}", "TransformationParameters": "{\"fillna\": \"mean\", \"transformations\": {\"0\": \"ClipOutliers\", \"1\": \"QuantileTransformer\"}, \"transformation_params\": {\"0\": {\"method\": \"clip\", \"std_threshold\": 1, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 1000}}}"}, "7fb9afb39d6fa9b428cf0f629c21bf27": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 5, \"point_method\": \"trimmed_mean_40\", \"distance_metric\": \"mae\", \"k\": 5, \"datepart_method\": [\"simple_binarized_poly\"], \"independent\": true}", "TransformationParameters": "{\"fillna\": \"rolling_mean\", \"transformations\": {\"0\": \"Constraint\", \"1\": \"QuantileTransformer\", \"2\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"constraint_method\": \"historic_diff\", \"constraint_direction\": \"lower\", \"constraint_regularization\": 1.0, \"constraint_value\": 0.2, \"bounds_only\": false, \"fillna\": null}, \"1\": {\"output_distribution\": \"uniform\", \"n_quantiles\": 43}, \"2\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 0.2, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"max\"}}}"}, "fd67d39f045cb25329dd5fedf940a41b": {"Model": "SeasonalityMotif", "ModelParameters": "{\"window\": 10, \"point_method\": \"median\", \"distance_metric\": \"mse\", \"k\": 10, \"datepart_method\": \"simple_binarized\", \"independent\": true}", "TransformationParameters": "{\"fillna\": \"ffill\", \"transformations\": {\"0\": \"AlignLastValue\", \"1\": \"AlignLastValue\"}, \"transformation_params\": {\"0\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}, \"1\": {\"rows\": 1, \"lag\": 1, \"method\": \"additive\", \"strength\": 1.0, \"first_value_only\": false, \"threshold\": null, \"threshold_method\": \"mean\"}}}"}}, "series": {"theta": "7fb9afb39d6fa9b428cf0f629c21bf27", "kappa": "7a14af550afa2194472cfc2e4e1440fb", "rho": "fd67d39f045cb25329dd5fedf940a41b", "sigma": "7a14af550afa2194472cfc2e4e1440fb"}} and transformations {} horizontal ensemble validations
Ensemble Horizontal component 1 of 3 LastValueNaive started
Ensemble Horizontal component 2 of 3 SeasonalityMotif started
Ensemble Horizontal component 3 of 3 SeasonalityMotif started
📈 1 - Ensemble with avg smape 30.99:
📈 2 - Ensemble with avg smape 30.84:
3 - Ensemble with avg smape 31.19:
📈 4 - Ensemble with avg smape 29.36:
📈 5 - Ensemble with avg smape 27.03:
6 - Ensemble with avg smape 31.13:
Template Eval Error: Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6596, in _fit
df = self._fit_one(df, i)
^^^^^^^^^^^^^^^^^^^^
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6574, in _fit_one
param=self.transformation_params[i],
~~~~~~~~~~~~~~~~~~~~~~~~~~^^^
KeyError: '2'
The above exception was the direct cause of the following exception:
Traceback (most recent call last):
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/evaluator/auto_model.py", line 2327, in TemplateWizard
transformer_object.fit(df_train)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6611, in fit
self._fit(df)
File "/home/runner/work/covid19-sir/covid19-sir/.venv/lib/python3.12/site-packages/autots/tools/transform.py", line 6601, in _fit
raise Exception(err_str) from e
Exception: Transformer HistoricValues failed on fit from params fake_date {'0': {'rows': 1, 'lag': 1, 'method': 'additive', 'strength': 1.0, 'first_value_only': False, 'threshold': 10, 'threshold_method': 'mean'}, '1': {'window': 28}} with error KeyError('2')
in model 6 in generation 0: Ensemble
Ensemble Horizontal component 1 of 3 LastValueNaive started
Ensemble Horizontal component 2 of 3 SeasonalityMotif started
Ensemble Horizontal component 3 of 3 SeasonalityMotif started
2025-11-22 at 12:42:49 | ERROR | validation of end failed
3. Compare scenarios
As explained with Tutorial: Scenario analysis, we can compare scenarios.
[14]:
# Adjust the last date, appending a phase
snr.append();
2025-11-22 at 12:42:49 | ERROR | validation of end failed
2025-11-22 at 12:42:49 | ERROR | validation of end failed
2025-11-22 at 12:42:49 | ERROR | validation of end failed
[15]:
# Compare reproduction number
ymin = snr.compare_param("Rt", date_range=(future_start_date, None), display=False).min().min()
snr.compare_param("Rt", date_range=(future_start_date, None), ylim=(ymin, None));
Note that minimum value of y in the figure was changed to focus on the differences of the scenarios.
[16]:
ymin_value = snr.compare_cases("Confirmed", date_range=(future_start_date, None), display=False).Predicted.min()
snr.compare_cases("Confirmed", date_range=(future_start_date, None), ylim=(ymin_value, None));
[17]:
# Show representative values
snr.describe()
[17]:
| max(Infected) | argmax(Infected) | Confirmed on 07Jul2023 | Infected on 07Jul2023 | Fatal on 07Jul2023 | |
|---|---|---|---|---|---|
| Baseline | 3716772.0 | 2022-08-15 | 34751309.0 | 282051.0 | 76456.0 |
| Predicted | 3716772.0 | 2022-08-15 | 34130443.0 | 54317.0 | 75521.0 |
| Predicted_with_X | 3716772.0 | 2022-08-15 | 34751309.0 | 282051.0 | 76456.0 |
Thank you!