{"id":"W4285787112","doi":"10.48550/arxiv.1909.02562","title":"TFCheck : A TensorFlow Library for Detecting Training Issues in Neural\\n Network Programs","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Implementation; Machine learning; Code (set theory); Artificial intelligence; Process (computing); Training set; Artificial neural network; Training (meteorology); Focus (optics); Software engineering; Programming language; Set (abstract data type)","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":[0.005794582,0.002008463,0.0007572367,0.002291071,0.001052434,0.001926668,0.002961899,0.001415174,0.01023161],"category_scores_gemma":[0.03009135,0.001236556,0.001586356,0.00102567,0.002440161,0.004683991,0.00266847,0.002599634,0.002275151],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002132535,"about_ca_system_score_gemma":0.003990563,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005111222,"about_ca_topic_score_gemma":0.005700021,"domain_scores_codex":[0.9968588,0.0008823129,0.0003646799,0.0005947045,0.001028261,0.0002711619],"domain_scores_gemma":[0.985665,0.008672519,0.001832758,0.002343525,0.001216535,0.0002697431],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001640246,0.0003959542,0.01994412,0.001800071,0.0003557165,0.001112159,0.001168959,0.2561147,0.03259233,0.08640822,0.08386649,0.514601],"study_design_scores_gemma":[0.00006841846,0.0001505803,0.0009189152,0.0001248808,0.00003540091,0.0002160141,0.00005054367,0.9064303,0.04008969,0.03881771,0.01303302,0.00006452136],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.01301546,0.0001927153,0.8522666,0.0003304707,0.0000984467,0.0001993241,0.001183358,0.1308783,0.001835459],"genre_scores_gemma":[0.2804747,0.0003892736,0.6928725,0.0005957525,0.0001100086,0.0008196387,0.004788396,0.01539705,0.00455278],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.01023161,"threshold_uncertainty_score":0.03422815,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09169102174530505,"score_gpt":0.2195692954900726,"score_spread":0.1278782737447676,"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."}}