{"id":"W4317035418","doi":"10.21203/rs.3.rs-2294878/v1","title":"(Almost) all the humpback whales of the North Pacific: A collaborative and comprehensive photo-ID dataset","year":2023,"lang":"en","type":"preprint","venue":"Research Square","topic":"Marine animal studies overview","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"Fisheries and Oceans Canada","funders":"Office of Naval Research; Fisheries and Oceans Canada; National Oceanic and Atmospheric Administration; Parks Canada; Strong; Consejo Nacional de Ciencia y Tecnología; U.S. Department of Defense","keywords":"Humpback whale; Geography; Scale (ratio); Resource (disambiguation); Marine mammal; Structural basin; Fishery; Citizen science; Sampling (signal processing); Identification (biology); Oceanography; Whale; Ecology; Cartography; Computer science; Geology; Biology","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0005333137,0.001077401,0.000649224,0.00363712,0.0007048657,0.001123886,0.001185984,0.00108114,0.0429284],"category_scores_gemma":[0.002079627,0.0005558205,0.0006725739,0.005924222,0.00032245,0.0008670453,0.001924021,0.0008865505,0.04739019],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006232538,"about_ca_system_score_gemma":0.001696736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.07119439,"about_ca_topic_score_gemma":0.1408475,"domain_scores_codex":[0.9994512,0.00003536655,0.00003531128,0.0001949748,0.0001593117,0.0001238343],"domain_scores_gemma":[0.9987574,0.0001715792,0.0001127254,0.0004114744,0.0003946953,0.0001521857],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.00007931997,0.00004599688,0.006946184,0.0004462703,0.00005261572,0.00004274297,0.0001595394,0.0003525222,0.001250427,0.0004321343,0.96883,0.02136237],"study_design_scores_gemma":[0.00004659827,0.0000131428,0.0627192,0.0001424526,0.00004055228,0.00009448751,0.0003774267,0.0004398343,0.00145913,0.0008779142,0.9337559,0.00003337006],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.001392637,0.00004888157,0.0002517988,0.00004822347,0.00001625247,0.0000244834,0.9956666,0.0004722911,0.002078819],"genre_scores_gemma":[0.002409174,0.00005885986,0.00117944,0.00003098787,0.000006798686,0.0001121351,0.9939109,0.000113593,0.00217828],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.07119439,"threshold_uncertainty_score":0.1436098,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1289366809424917,"score_gpt":0.3822911479379739,"score_spread":0.2533544669954821,"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."}}