{"id":"W2983385682","doi":"10.1016/j.envsoft.2020.104926","title":"The proper care and feeding of CAMELS: How limited training data affects streamflow prediction","year":2020,"lang":"en","type":"article","venue":"Environmental Modelling & Software","topic":"Hydrology and Watershed Management Studies","field":"Environmental Science","cited_by":243,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo","funders":"Canada First Research Excellence Fund","keywords":"Streamflow; Training (meteorology); Computer science; Training set; Tree (set theory); Set (abstract data type); Machine learning; Data set; Sequence (biology); Predictive modelling; Data mining; Artificial intelligence; Mathematics; Meteorology; Geography; Drainage basin; Cartography","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.003475936,0.0002475303,0.0003039324,0.000226486,0.0003875201,0.001001683,0.000864327,0.00101547,0.001586526],"category_scores_gemma":[0.02970827,0.0001709223,0.0002912823,0.000243014,0.0006139413,0.001170933,0.0004594013,0.0006466768,0.0004128455],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008923771,"about_ca_system_score_gemma":0.000755928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01933216,"about_ca_topic_score_gemma":0.02034022,"domain_scores_codex":[0.9987645,0.0006351509,0.00006177303,0.0003027982,0.0001343525,0.0001014836],"domain_scores_gemma":[0.9848152,0.01110451,0.0008639095,0.001157936,0.001426461,0.000631872],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00491575,0.0009728592,0.8098484,0.0002396263,0.0003378098,0.0007470116,0.001402111,0.03625954,0.02117406,0.0008818853,0.006082554,0.1171384],"study_design_scores_gemma":[0.0001700234,0.001992252,0.7812738,0.0002506708,0.0004671593,0.0005366914,0.001650404,0.1826547,0.02322637,0.001778055,0.005885912,0.0001138258],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9960302,0.00008536852,0.001751955,0.0003493144,0.00002730958,0.00001643648,0.0004221309,0.00007961843,0.001237681],"genre_scores_gemma":[0.9967379,0.00004829617,0.001715158,0.0001504709,0.000006098845,0.00002259492,0.0007211343,0.00003502204,0.0005633784],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.996524,"threshold_uncertainty_score":0.03843927,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04101356804679195,"score_gpt":0.1996113489572896,"score_spread":0.1585977809104976,"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."}}