[ { "trial_number": 0, "params": { "num_heads": 6, "num_layers": 1 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.1832761391537991, "val_loss": 0.7498329176104127, "macro_f1": 0.7103340867858469 }, { "epoch": 2, "train_loss": 0.6320729479836771, "val_loss": 0.586862810580014, "macro_f1": 0.7913692983317114 }, { "epoch": 3, "train_loss": 0.5160054902002932, "val_loss": 0.5864161346933524, "macro_f1": 0.7848025166573419 }, { "epoch": 4, "train_loss": 0.411905202302627, "val_loss": 0.511089034415231, "macro_f1": 0.8072666472893073 }, { "epoch": 5, "train_loss": 0.3297956578902046, "val_loss": 0.464968596516278, "macro_f1": 0.834067964116802 }, { "epoch": 6, "train_loss": 0.2837876489732768, "val_loss": 0.38831818485494907, "macro_f1": 0.8705889067805529 }, { "epoch": 7, "train_loss": 0.22080586585141582, "val_loss": 0.351857351157539, "macro_f1": 0.8878972081948868 }, { "epoch": 8, "train_loss": 0.18440273702695836, "val_loss": 0.36548399620602284, "macro_f1": 0.8804424351677086 }, { "epoch": 9, "train_loss": 0.14913206719753186, "val_loss": 0.38098949995442416, "macro_f1": 0.8723550071229064 }, { "epoch": 10, "train_loss": 0.12198060535528363, "val_loss": 0.4393947260646984, "macro_f1": 0.8655638957350722 } ], "final_macro_f1": 0.8655638957350722 }, { "trial_number": 1, "params": { "num_heads": 4, "num_layers": 2 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.1230820458561512, "val_loss": 0.7704554694920338, "macro_f1": 0.7036616295856207 }, { "epoch": 2, "train_loss": 0.6215897324493869, "val_loss": 0.6329705550752837, "macro_f1": 0.747321270691553 }, { "epoch": 3, "train_loss": 0.5201769666298397, "val_loss": 0.5488664073309875, "macro_f1": 0.8103063427800269 }, { "epoch": 4, "train_loss": 0.3938881587673498, "val_loss": 0.506728891118113, "macro_f1": 0.8111035470961853 }, { "epoch": 5, "train_loss": 0.3405428771871233, "val_loss": 0.4811670455824712, "macro_f1": 0.8425270143911638 }, { "epoch": 6, "train_loss": 0.29019304942776036, "val_loss": 0.41828343503434084, "macro_f1": 0.8464476111160173 }, { "epoch": 7, "train_loss": 0.23187170347196548, "val_loss": 0.42985851424080984, "macro_f1": 0.8487845567154556 }, { "epoch": 8, "train_loss": 0.19625346978113478, "val_loss": 0.42691103950730097, "macro_f1": 0.8671223625188696 }, { "epoch": 9, "train_loss": 0.14775698915877267, "val_loss": 0.44993914651175215, "macro_f1": 0.8621576111982389 }, { "epoch": 10, "train_loss": 0.1359217370229827, "val_loss": 0.40257360588924346, "macro_f1": 0.8679347055481997 } ], "final_macro_f1": 0.8679347055481997 }, { "trial_number": 4, "params": { "num_heads": 6, "num_layers": 2 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.3173663898402166, "val_loss": 0.759101252397293, "macro_f1": 0.6990005637236743 }, { "epoch": 2, "train_loss": 0.6467822624046617, "val_loss": 0.5983039810152476, "macro_f1": 0.78449167857198 }, { "epoch": 3, "train_loss": 0.5089347752778068, "val_loss": 0.5390065236044634, "macro_f1": 0.8095188828227796 }, { "epoch": 4, "train_loss": 0.40332894629894145, "val_loss": 0.4772564168340467, "macro_f1": 0.8295270726746916 }, { "epoch": 5, "train_loss": 0.33736765722752204, "val_loss": 0.46593107330828465, "macro_f1": 0.8366828266689496 }, { "epoch": 6, "train_loss": 0.26871140042156827, "val_loss": 0.46045689791294153, "macro_f1": 0.8419257227035141 }, { "epoch": 7, "train_loss": 0.21393143714539956, "val_loss": 0.39201908384270856, "macro_f1": 0.8753177666423365 }, { "epoch": 8, "train_loss": 0.1685226441265696, "val_loss": 0.4256001700615061, "macro_f1": 0.8579319045518754 }, { "epoch": 9, "train_loss": 0.11896324822100388, "val_loss": 0.43303222099108063, "macro_f1": 0.8677495291666183 }, { "epoch": 10, "train_loss": 0.1051060429541107, "val_loss": 0.49004084796741093, "macro_f1": 0.8619553747771226 } ], "final_macro_f1": 0.8619553747771226 }, { "trial_number": 5, "params": { "num_heads": 6, "num_layers": 2 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.1954575238701153, "val_loss": 0.7813598942286862, "macro_f1": 0.7221217333984522 }, { "epoch": 2, "train_loss": 0.6292221567486718, "val_loss": 0.6820206618661364, "macro_f1": 0.730189742107776 }, { "epoch": 3, "train_loss": 0.5232538815344306, "val_loss": 0.5693920787625713, "macro_f1": 0.7875359818027032 }, { "epoch": 4, "train_loss": 0.4423905862759427, "val_loss": 0.5050926569647389, "macro_f1": 0.8160427418514019 }, { "epoch": 5, "train_loss": 0.35481051066212826, "val_loss": 0.45971177048577466, "macro_f1": 0.8303637056003775 }, { "epoch": 6, "train_loss": 0.29134497639944756, "val_loss": 0.4631793054160226, "macro_f1": 0.8430694304621142 }, { "epoch": 7, "train_loss": 0.2574463309639773, "val_loss": 0.4040321405481529, "macro_f1": 0.8648266468143123 }, { "epoch": 8, "train_loss": 0.17930574953997033, "val_loss": 0.41403597299688555, "macro_f1": 0.8802919260037383 }, { "epoch": 9, "train_loss": 0.15790989192182017, "val_loss": 0.44370942524281043, "macro_f1": 0.8783024228283974 }, { "epoch": 10, "train_loss": 0.13971920217285175, "val_loss": 0.4138593869329673, "macro_f1": 0.8782744869992875 } ], "final_macro_f1": 0.8782744869992875 }, { "trial_number": 7, "params": { "num_heads": 4, "num_layers": 2 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.1595548766902872, "val_loss": 0.7383359403445803, "macro_f1": 0.712648753378513 }, { "epoch": 2, "train_loss": 0.6291670470364144, "val_loss": 0.6041549987393647, "macro_f1": 0.7628297499692137 }, { "epoch": 3, "train_loss": 0.5113105823535072, "val_loss": 0.5913394728464446, "macro_f1": 0.762223321208065 }, { "epoch": 4, "train_loss": 0.429566226302475, "val_loss": 0.5100458779285106, "macro_f1": 0.8092755734750158 }, { "epoch": 5, "train_loss": 0.34133534183043585, "val_loss": 0.4490610392663279, "macro_f1": 0.8514615846241297 }, { "epoch": 6, "train_loss": 0.28462050195539923, "val_loss": 0.4344335053795077, "macro_f1": 0.8501548126586678 }, { "epoch": 7, "train_loss": 0.2287863478266649, "val_loss": 0.43782980841001856, "macro_f1": 0.8551092733111149 }, { "epoch": 8, "train_loss": 0.17019800981449287, "val_loss": 0.4136471379493258, "macro_f1": 0.8771639159798625 }, { "epoch": 9, "train_loss": 0.13897078628760892, "val_loss": 0.3903981795387682, "macro_f1": 0.8690399950721789 }, { "epoch": 10, "train_loss": 0.1324176407532833, "val_loss": 0.39504132581315016, "macro_f1": 0.8887459647104133 } ], "final_macro_f1": 0.8887459647104133 }, { "trial_number": 8, "params": { "num_heads": 8, "num_layers": 3 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.2111919544334036, "val_loss": 0.7954880866804733, "macro_f1": 0.692173249027094 }, { "epoch": 2, "train_loss": 0.6736151260012616, "val_loss": 0.6812303161679818, "macro_f1": 0.7403447069975888 }, { "epoch": 3, "train_loss": 0.5472460934655734, "val_loss": 0.5812115360950602, "macro_f1": 0.7688338387674025 }, { "epoch": 4, "train_loss": 0.46018385821918967, "val_loss": 0.5203743778132452, "macro_f1": 0.8032102618967811 }, { "epoch": 5, "train_loss": 0.37938465003461636, "val_loss": 0.5028770503409097, "macro_f1": 0.8164099342185992 }, { "epoch": 6, "train_loss": 0.316088581005279, "val_loss": 0.41271519390269745, "macro_f1": 0.8469800383056026 }, { "epoch": 7, "train_loss": 0.25914997950141144, "val_loss": 0.44095426879529587, "macro_f1": 0.8534820881525831 }, { "epoch": 8, "train_loss": 0.20093080279694092, "val_loss": 0.4070406252233853, "macro_f1": 0.8711430333008884 }, { "epoch": 9, "train_loss": 0.18884633775942863, "val_loss": 0.4072782529949261, "macro_f1": 0.8589692051642381 }, { "epoch": 10, "train_loss": 0.13745949061360077, "val_loss": 0.38686535329434085, "macro_f1": 0.8790647966307543 } ], "final_macro_f1": 0.8790647966307543 }, { "trial_number": 9, "params": { "num_heads": 6, "num_layers": 1 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.1922890041230199, "val_loss": 0.717364141946943, "macro_f1": 0.7111832146234033 }, { "epoch": 2, "train_loss": 0.614923968415078, "val_loss": 0.6068004227330532, "macro_f1": 0.7572401128403796 }, { "epoch": 3, "train_loss": 0.5134530554166117, "val_loss": 0.5359148223118242, "macro_f1": 0.8108260298969783 }, { "epoch": 4, "train_loss": 0.4162116786775107, "val_loss": 0.5204935171113813, "macro_f1": 0.8078057743572108 }, { "epoch": 5, "train_loss": 0.33961279247456794, "val_loss": 0.5185712169984291, "macro_f1": 0.8156870159488682 }, { "epoch": 6, "train_loss": 0.2688734014731006, "val_loss": 0.5058216502422036, "macro_f1": 0.8127487330103519 }, { "epoch": 7, "train_loss": 0.24128333852154407, "val_loss": 0.44323977637173506, "macro_f1": 0.8459015831040886 }, { "epoch": 8, "train_loss": 0.19852170058779886, "val_loss": 0.3472442598325278, "macro_f1": 0.8831003044387479 }, { "epoch": 9, "train_loss": 0.12439552349590785, "val_loss": 0.3821052959179643, "macro_f1": 0.8758038683098553 }, { "epoch": 10, "train_loss": 0.14608042874935045, "val_loss": 0.3752268017476122, "macro_f1": 0.8712872641014485 } ], "final_macro_f1": 0.8712872641014485 }, { "trial_number": 11, "params": { "num_heads": 8, "num_layers": 3 }, "epoch_metrics": [ { "epoch": 1, "train_loss": 1.3417097340382718, "val_loss": 0.7926707279505988, "macro_f1": 0.7129917353338494 }, { "epoch": 2, "train_loss": 0.6865363907431851, "val_loss": 0.6855129791891633, "macro_f1": 0.7429053146243679 }, { "epoch": 3, "train_loss": 0.5470005607854864, "val_loss": 0.6210718623229435, "macro_f1": 0.763000265572399 }, { "epoch": 4, "train_loss": 0.4693222763946994, "val_loss": 0.5969800439300795, "macro_f1": 0.7693955538580185 }, { "epoch": 5, "train_loss": 0.36977021348388217, "val_loss": 0.5299910488974284, "macro_f1": 0.8096957303892384 }, { "epoch": 6, "train_loss": 0.31427258129992997, "val_loss": 0.525438177046077, "macro_f1": 0.830292606084595 }, { "epoch": 7, "train_loss": 0.278600519309429, "val_loss": 0.4611006113402362, "macro_f1": 0.8491567596313283 }, { "epoch": 8, "train_loss": 0.21794261691492378, "val_loss": 0.5124526821774215, "macro_f1": 0.8308818032791065 }, { "epoch": 9, "train_loss": 0.2131353972018059, "val_loss": 0.4707893036269202, "macro_f1": 0.8478198878118651 }, { "epoch": 10, "train_loss": 0.15050037305815886, "val_loss": 0.4796101963101643, "macro_f1": 0.8644901534658801 } ], "final_macro_f1": 0.8644901534658801 } ]