task: pubmed_qa dataset_path: pubmed_qa dataset_name: pqa_labeled output_type: multiple_choice training_split: null validation_split: null test_split: train doc_to_text: !function preprocess_pubmedqa.doc_to_text doc_to_target: final_decision doc_to_choice: ["yes", "no", "maybe"] metric_list: - metric: acc aggregation: mean higher_is_better: true