{"id":"W4398652322","doi":"10.7910/dvn/k0oyqf/fhu04w","title":"preprocess.py","year":2019,"lang":"en","type":"dataset","venue":"Harvard Dataverse","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001767819,0.00542077,0.001943474,0.006030023,0.001123722,0.004251209,0.004103578,0.002492308,0.178931],"category_scores_gemma":[0.007527263,0.001319567,0.002414385,0.006732239,0.0008455958,0.002775398,0.003377325,0.002277611,0.3318375],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001641738,"about_ca_system_score_gemma":0.002378371,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01266091,"about_ca_topic_score_gemma":0.02490941,"domain_scores_codex":[0.9986244,0.0001829916,0.0001528377,0.0005309319,0.0002592778,0.0002494977],"domain_scores_gemma":[0.9970949,0.0005701822,0.0001501895,0.001376094,0.0005684973,0.0002401139],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00006331894,0.00002089376,0.0002244965,0.0003956298,0.00003242678,0.00001159986,0.0000155327,0.0001274611,0.0001830685,0.0002170868,0.995002,0.003706459],"study_design_scores_gemma":[0.0002511269,0.00005067489,0.001504558,0.0001464214,0.00003777178,0.00006537886,0.00005459082,0.0009678222,0.001721073,0.002378726,0.9927759,0.00004597736],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.0003261105,0.0002405384,0.0005167561,0.0001803222,0.0001638609,0.00005652776,0.9798844,0.01594085,0.002690656],"genre_scores_gemma":[0.0009551014,0.0001493527,0.001278044,0.0001314127,0.00003768342,0.0001295806,0.9933583,0.001601174,0.002359303],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.821069,"threshold_uncertainty_score":0.5985837,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02798996857792456,"score_gpt":0.2739853967227555,"score_spread":0.2459954281448309,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}