{"id":"W3160700180","doi":"10.1109/icse43902.2021.00138","title":"Automatic Unit Test Generation for Machine Learning Libraries: How Far Are We?","year":2021,"lang":"en","type":"article","venue":"","topic":"Software Testing and Debugging Techniques","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"Natural Sciences and Engineering Research Council of Canada; National Natural Science Foundation of China","keywords":"Unit testing; Test suite; Machine learning; Computer science; Artificial intelligence; Test (biology); Test Management Approach; Code coverage; Unit (ring theory); Test set; Set (abstract data type); Software; Quality (philosophy); Test case; Software development; Programming language; Software construction","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001737339,0.0001116482,0.0001311013,0.00006350347,0.0002298965,0.00056422,0.0003128591,0.00005265874,0.00002402179],"category_scores_gemma":[0.001843655,0.00009983285,0.00004594424,0.0003305615,0.00001634206,0.0005349256,0.0001654691,0.0001041039,0.000006466459],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001396166,"about_ca_system_score_gemma":0.00008650944,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001110867,"about_ca_topic_score_gemma":0.0000104561,"domain_scores_codex":[0.9992205,0.00005071065,0.0001230805,0.0002941416,0.0001327241,0.0001788452],"domain_scores_gemma":[0.9985994,0.0007661906,0.00008311722,0.0003679812,0.0001313043,0.00005202819],"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":[0.000002481868,0.0003630808,0.07053103,0.0004149179,0.00006592466,0.0001179224,0.001786721,0.0005762675,0.006387762,0.0871675,0.2927419,0.5398445],"study_design_scores_gemma":[0.0001136822,0.00007176192,0.0003075828,0.00004539838,0.000004676232,0.00002616136,0.00001288738,0.966528,0.006860455,0.01310586,0.01277438,0.0001491647],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.001284585,0.0003545523,0.9755098,0.009056669,0.000108591,0.0001054362,0.000002568252,0.01335205,0.0002257567],"genre_scores_gemma":[0.1566517,0.0000191516,0.8393217,0.0008609892,0.0000852755,0.00003931085,0.0000243976,0.00001480353,0.002982743],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9659517,"threshold_uncertainty_score":0.5440786,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06743097283882477,"score_gpt":0.2686211446901192,"score_spread":0.2011901718512945,"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."}}