auto-sklearn: Error creating dummy predictions: {'error': 'Result queue is empty', 'configuration_origin': 'DUMMY'}

Greetings,

I am trying to run autosklearn 0.6.0 on an HPC infrastructure. I am using the kr-vs-kp dataset (https://www.openml.org/d/3) and I have made all the values of the dataset numeric. The error I get is:

Unhandled exception in thread started by libgcc_s.so.1 must be installed for pthread_cancel to work Traceback (most recent call last): File "main_script.py", line 41, in <module> autosklearn_script.autosklearn_func(train_data,train_target,test_data,test_target,dataset_info_dict,task,time_threshold) File "/users/pr004/ixanthos/test_files/autosklearn_script.py", line 31, in autosklearn_func automl.fit(train_data, train_target, metric=autosklearn.metrics.roc_auc) File "/apps/applications/pytorch/1.3.1/lib/python3.6/site-packages/autosklearn/estimators.py", line 664, in fit dataset_name=dataset_name, File "/apps/applications/pytorch/1.3.1/lib/python3.6/site-packages/autosklearn/estimators.py", line 337, in fit self._automl[0].fit(**kwargs) File "/apps/applications/pytorch/1.3.1/lib/python3.6/site-packages/autosklearn/automl.py", line 996, in fit load_models=load_models, File "/apps/applications/pytorch/1.3.1/lib/python3.6/site-packages/autosklearn/automl.py", line 208, in fit only_return_configuration_space=only_return_configuration_space, File "/apps/applications/pytorch/1.3.1/lib/python3.6/site-packages/autosklearn/automl.py", line 384, in _fit num_run = self._do_dummy_prediction(datamanager, num_run) File "/apps/applications/pytorch/1.3.1/lib/python3.6/site-packages/autosklearn/automl.py", line 313, in _do_dummy_prediction raise ValueError("Dummy prediction failed: %s " % str(additional_info)) ValueError: Dummy prediction failed: {'error': 'Result queue is empty', 'configuration_origin': 'DUMMY'}

Any ideas?

Thank you in advance.

About this issue

  • Original URL
  • State: closed
  • Created 4 years ago
  • Reactions: 7
  • Comments: 17 (11 by maintainers)

Commits related to this issue

Most upvoted comments

Hey, this issue is getting hard to follow through as we have changed Auto-sklearn quite a bit since most answers.

@MaxHeuillet: if this is still relevant for you could you please open a new issue as your issue appears to be related to a special hardware platform. Not sure if we can help, but it’ll make it easier to keep track of things. @rabsr: We have improved the printed output quite a bit. Could you please open a new issue if you are still experiencing this? @sofidenner: As your issue already has extra output that we have added quite some time ago it would be great if you could open a new issue with what you already posted and also a) whether it happens repeatedly or only stochastically, b) additional info on the dataset you’re using.