{"id":"W3139790935","doi":"10.1002/rse2.200","title":"21 000 birds in 4.5 h: efficient large‐scale seabird detection with machine learning","year":2021,"lang":"en","type":"article","venue":"Remote Sensing in Ecology and Conservation","topic":"Remote Sensing and LiDAR Applications","field":"Environmental Science","cited_by":61,"is_retracted":false,"has_abstract":true,"ca_institutions":"Squamish Nation; Quest University Canada","funders":"MAVA Foundation; Nvidia","keywords":"Seabird; Convolutional neural network; Artificial intelligence; Predation; Bioindicator; Computer science; Precision and recall; Deep learning; Ecology; Machine learning; Biology","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":[],"consensus_categories":[],"category_scores_codex":[0.0003644075,0.0008831078,0.0004937059,0.0007214191,0.000221702,0.0004845878,0.001070827,0.0005927111,0.002666731],"category_scores_gemma":[0.001037445,0.000358212,0.0003539342,0.0004254559,0.0002128533,0.0008692933,0.0006261049,0.0004806286,0.00165069],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005072796,"about_ca_system_score_gemma":0.0005365967,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01217071,"about_ca_topic_score_gemma":0.02027021,"domain_scores_codex":[0.9997374,0.0000311183,0.00001038042,0.00008863079,0.00007700516,0.00005545717],"domain_scores_gemma":[0.9996196,0.0001295955,0.00004993111,0.00007412844,0.00008573272,0.00004105342],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.001211035,0.0005953752,0.02586194,0.0002453935,0.0002225482,0.0004219821,0.0001101495,0.1458789,0.05714974,0.0006222946,0.02488074,0.7427999],"study_design_scores_gemma":[0.00002190637,0.00007622498,0.007380412,0.00001034881,0.00001782248,0.00006677303,0.0000407349,0.9794599,0.01019881,0.0006963416,0.002016706,0.00001394622],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6985135,0.001716521,0.2644776,0.0008203973,0.0005280981,0.0001834441,0.003134213,0.02511852,0.005507679],"genre_scores_gemma":[0.9189576,0.0001581169,0.07549243,0.0001957963,0.00007579783,0.00006903841,0.002345752,0.0001601152,0.002545377],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01217071,"threshold_uncertainty_score":0.02419972,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.006278993618221449,"score_gpt":0.2088509544687405,"score_spread":0.202571960850519,"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."}}