Description

My understanding is that in order to upload the results for a particular task, I need to create a run with a trained model and task such as follows:

task = openml.tasks.get_task(119) 
clf = ensemble.RandomForestClassifier() 
run = openml.runs.run_model_on_task(clf, task)

Would it be possible to upload only the prediction (e.g. as a csv file) for a given task? Instead of needing to upload a trained model. Like it is done in Kaggle competitions or challenges on https://eval.ai/.