-
Notifications
You must be signed in to change notification settings - Fork 303
Description
After normal download and installation of autopytorch 0.2.1, even the official example file example_time_series_forecasting.py cannot run, with the following error:
python example_time_series_forecasting.py
Traceback (most recent call last):
File "example_time_series_forecasting.py", line 57, in
api.search(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/api/time_series_forecasting.py", line 476, in search
return self._search(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/api/base_task.py", line 1011, in _search
dataset_requirements = get_dataset_requirements(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/utils/pipeline.py", line 83, in get_dataset_requirements
return _get_forecasting_dataset_requirements(info,
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/utils/pipeline.py", line 141, in _get_forecasting_dataset_requirements
return TimeSeriesForecastingPipeline(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/pipeline/time_series_forecasting.py", line 91, in init
BasePipeline.init(self,
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/pipeline/base_pipeline.py", line 104, in init
self.config_space = self.get_hyperparameter_search_space()
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/pipeline/base_pipeline.py", line 262, in get_hyperparameter_search_space
self.config_space = self._get_hyperparameter_search_space(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/pipeline/time_series_forecasting.py", line 161, in _get_hyperparameter_search_space
cs = self._get_base_search_space(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/pipeline/base_pipeline.py", line 376, in _get_base_search_space
sub_config_space = node.get_hyperparameter_search_space( # type: ignore[call-arg]
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/autoPyTorch/pipeline/components/setup/network_backbone/forecasting_backbone/init.py", line 206, in get_hyperparameter_search_space
hp_encoder = CSH.CategoricalHyperparameter(
File "/home/funmatlab/opt/anaconda3/envs/auto-pytorch/lib/python3.8/site-packages/ConfigSpace/hyperparameters/categorical.py", line 219, in init
raise ValueError(
ValueError: The default value has to be one of the choices. Got None which is not in ('flat_encoder', 'seq_encoder').