{"id":"W2972233065","doi":"10.1109/qrs.2019.00059","title":"TFCheck : A TensorFlow Library for Detecting Training Issues in Neural Network Programs","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Computer science; Implementation; Machine learning; Code (set theory); Training set; Artificial intelligence; Training (meteorology); Process (computing); Artificial neural network; Focus (optics); Software engineering; Data mining; 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":[0.007551754,0.002482144,0.00101805,0.002987237,0.001171271,0.002336311,0.003851538,0.00172423,0.008617418],"category_scores_gemma":[0.03806082,0.0016445,0.001927216,0.001356235,0.002852397,0.005505492,0.002870595,0.002833723,0.001975981],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002267839,"about_ca_system_score_gemma":0.004083487,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004937432,"about_ca_topic_score_gemma":0.004877382,"domain_scores_codex":[0.9957631,0.001183594,0.000555779,0.0008028719,0.001346185,0.0003484868],"domain_scores_gemma":[0.9777954,0.01337397,0.003019159,0.003704301,0.001733822,0.0003733125],"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":[0.00240164,0.0005767819,0.0299749,0.002660291,0.0005263722,0.001547746,0.00177999,0.2659984,0.04709596,0.06687333,0.07649351,0.5040711],"study_design_scores_gemma":[0.00009815578,0.00019649,0.001356506,0.0001578406,0.00005267989,0.0002647695,0.00006620796,0.8995711,0.04990293,0.03752202,0.01072143,0.00008992501],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.01915651,0.0002222482,0.7895868,0.0003115473,0.000108405,0.0002333096,0.001362669,0.1876628,0.00135571],"genre_scores_gemma":[0.2996633,0.0003981225,0.6683045,0.0005908904,0.0001118226,0.0009027781,0.004788728,0.02172622,0.003513665],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.008617418,"threshold_uncertainty_score":0.03993797,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04738158154081169,"score_gpt":0.2962145034165068,"score_spread":0.2488329218756952,"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."}}