search

intive-DataScience / tbats

BATS and TBATS forecasting methods
MIT License
178 stars 19 forks source link

Input contains NaN on multiprocessing/pool.py #8

Closed LoveenDyall closed 4 years ago

LoveenDyall commented 5 years ago

File "../forecasters/statsforecasters.py", line 78, in search model_fit = model.fit(train) File "../.local/lib/python3.5/site-packages/tbats/abstract/Estimator.py", line 97, in fit best_model = self._do_fit(y) File "../.local/lib/python3.5/site-packages/tbats/tbats/TBATS.py", line 76, in _do_fit seasonal_model = self._choose_model_from_possible_component_settings(y, components_grid=components_grid) File "../.local/lib/python3.5/site-packages/tbats/abstract/Estimator.py", line 143, in _choose_model_from_possible_component_settings models = pool.map(self._case_fit, components_grid) File "/usr/lib/python3.5/multiprocessing/pool.py", line 260, in map return self._map_async(func, iterable, mapstar, chunksize).get() File "/usr/lib/python3.5/multiprocessing/pool.py", line 608, in get raise self._value ValueError: Input contains NaN, infinity or a value too large for dtype('float64')

Hi I also keep getting this error, I have verified my training data does not contain any NaN or infinite values using: if np.isfinite(train).all():

also i'm determining seasonal_periods using pacf from statsmodels python package, and picking out the lag with highest confidence coefficient

cotterpl commented 5 years ago

Can you try turning box cox off using use_box_cox=False? Does it solve the problem?

cotterpl commented 4 years ago

Closing as I am unable to reproduce this and there has been no feedback on this for a few months.