{"id":"W4390951092","doi":"10.53055/icimod.1016","title":"Exploring the potential of deep learning for classifying camera trap data: A case study from Nepal - working paper","year":2023,"lang":"en","type":"report","venue":"","topic":"Wildlife Ecology and Conservation","field":"Environmental Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"United Nations University Institute for Water, Environment, and Health","funders":"","keywords":"Camera trap; Scope (computer science); Transfer of learning; Computer science; Wildlife; Deep learning; Artificial intelligence; Trap (plumbing); Key (lock); Data science; Machine learning; Geography; Ecology; Computer security","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":[0.0009673209,0.0004170016,0.0002054339,0.0009065822,0.0006364168,0.001105954,0.0009443676,0.0008983159,0.0008381347],"category_scores_gemma":[0.003081937,0.000165946,0.0002499588,0.0009083156,0.0005919596,0.0008588376,0.0007083019,0.0008474111,0.0003003476],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009777937,"about_ca_system_score_gemma":0.0005779567,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.05053667,"about_ca_topic_score_gemma":0.1072235,"domain_scores_codex":[0.9995383,0.0001737695,0.00002882411,0.00007129996,0.0001102355,0.00007755992],"domain_scores_gemma":[0.9979006,0.001412481,0.000098899,0.0001825372,0.0003300112,0.00007550182],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007522891,0.002098671,0.3577924,0.0009567655,0.0003746603,0.01861796,0.01060107,0.09601185,0.01494705,0.005711452,0.0227954,0.4693404],"study_design_scores_gemma":[0.0001230187,0.0007420583,0.2869346,0.0004178385,0.0001950906,0.006307607,0.02537222,0.6050968,0.02927309,0.0089115,0.03644907,0.0001770752],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9850993,0.0002021483,0.00579634,0.00114767,0.00001520188,0.00008837599,0.0008923152,0.0001468213,0.006611981],"genre_scores_gemma":[0.9834646,0.0001541838,0.01208089,0.0001901529,0.00001027025,0.00004272123,0.001261084,0.00002585935,0.002770198],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.05053667,"threshold_uncertainty_score":0.100485,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2824870302698901,"score_gpt":0.3334506354115269,"score_spread":0.05096360514163678,"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."}}