| { | |
| "best_metric": 0.1773720234632492, | |
| "best_model_checkpoint": "wav2vec2-xlsr-1b-mecita-portuguese-all-text-a_coisa-os_morcegos/checkpoint-1738", | |
| "epoch": 42.0, | |
| "global_step": 3318, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 3e-05, | |
| "loss": 25.5905, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.0734213636130093, | |
| "eval_loss": 0.44948670268058777, | |
| "eval_runtime": 19.9817, | |
| "eval_samples_per_second": 15.614, | |
| "eval_steps_per_second": 1.952, | |
| "eval_wer": 0.25804706518799025, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 2.9631645569620254e-05, | |
| "loss": 3.1482, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.038020860093510765, | |
| "eval_loss": 0.2479224056005478, | |
| "eval_runtime": 19.8794, | |
| "eval_samples_per_second": 15.695, | |
| "eval_steps_per_second": 1.962, | |
| "eval_wer": 0.12036786583716527, | |
| "step": 158 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 2.9251898734177216e-05, | |
| "loss": 0.4247, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.03447567178749422, | |
| "eval_loss": 0.23473307490348816, | |
| "eval_runtime": 19.9054, | |
| "eval_samples_per_second": 15.674, | |
| "eval_steps_per_second": 1.959, | |
| "eval_wer": 0.10251555315120368, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 2.887215189873418e-05, | |
| "loss": 0.3136, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 0.032214971998150335, | |
| "eval_loss": 0.2043968290090561, | |
| "eval_runtime": 20.064, | |
| "eval_samples_per_second": 15.55, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.10170408439275087, | |
| "step": 316 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 0.029594615424138108, | |
| "eval_loss": 0.19060839712619781, | |
| "eval_runtime": 19.9461, | |
| "eval_samples_per_second": 15.642, | |
| "eval_steps_per_second": 1.955, | |
| "eval_wer": 0.09304841763592102, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 2.849240506329114e-05, | |
| "loss": 0.2985, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.031084622103478395, | |
| "eval_loss": 0.2050233781337738, | |
| "eval_runtime": 20.0744, | |
| "eval_samples_per_second": 15.542, | |
| "eval_steps_per_second": 1.943, | |
| "eval_wer": 0.09629429266973222, | |
| "step": 474 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 2.8112658227848103e-05, | |
| "loss": 0.2413, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.030879103940810768, | |
| "eval_loss": 0.20249603688716888, | |
| "eval_runtime": 20.1379, | |
| "eval_samples_per_second": 15.493, | |
| "eval_steps_per_second": 1.937, | |
| "eval_wer": 0.09710576142818501, | |
| "step": 553 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 2.7732911392405062e-05, | |
| "loss": 0.2267, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.02913219955813595, | |
| "eval_loss": 0.20055869221687317, | |
| "eval_runtime": 19.8715, | |
| "eval_samples_per_second": 15.701, | |
| "eval_steps_per_second": 1.963, | |
| "eval_wer": 0.08845009467135516, | |
| "step": 632 | |
| }, | |
| { | |
| "epoch": 8.86, | |
| "learning_rate": 2.7353164556962024e-05, | |
| "loss": 0.224, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.029080820017469044, | |
| "eval_loss": 0.1990896761417389, | |
| "eval_runtime": 19.8842, | |
| "eval_samples_per_second": 15.691, | |
| "eval_steps_per_second": 1.961, | |
| "eval_wer": 0.09169596970516636, | |
| "step": 711 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.02810460874479782, | |
| "eval_loss": 0.1880808174610138, | |
| "eval_runtime": 19.8781, | |
| "eval_samples_per_second": 15.696, | |
| "eval_steps_per_second": 1.962, | |
| "eval_wer": 0.08845009467135516, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 10.13, | |
| "learning_rate": 2.697341772151899e-05, | |
| "loss": 0.1864, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.02784771104146329, | |
| "eval_loss": 0.18414445221424103, | |
| "eval_runtime": 20.0151, | |
| "eval_samples_per_second": 15.588, | |
| "eval_steps_per_second": 1.949, | |
| "eval_wer": 0.08926156342980796, | |
| "step": 869 | |
| }, | |
| { | |
| "epoch": 11.39, | |
| "learning_rate": 2.659367088607595e-05, | |
| "loss": 0.1951, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_cer": 0.028155988285464728, | |
| "eval_loss": 0.18085478246212006, | |
| "eval_runtime": 20.0956, | |
| "eval_samples_per_second": 15.526, | |
| "eval_steps_per_second": 1.941, | |
| "eval_wer": 0.08953205301595889, | |
| "step": 948 | |
| }, | |
| { | |
| "epoch": 12.66, | |
| "learning_rate": 2.621392405063291e-05, | |
| "loss": 0.1794, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_cer": 0.028001849663464008, | |
| "eval_loss": 0.19231769442558289, | |
| "eval_runtime": 20.0602, | |
| "eval_samples_per_second": 15.553, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.08331079253448742, | |
| "step": 1027 | |
| }, | |
| { | |
| "epoch": 13.92, | |
| "learning_rate": 2.5834177215189873e-05, | |
| "loss": 0.1621, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_cer": 0.027744951960129476, | |
| "eval_loss": 0.19489158689975739, | |
| "eval_runtime": 20.0375, | |
| "eval_samples_per_second": 15.571, | |
| "eval_steps_per_second": 1.946, | |
| "eval_wer": 0.08574519880984582, | |
| "step": 1106 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_cer": 0.02656322252479063, | |
| "eval_loss": 0.19287976622581482, | |
| "eval_runtime": 20.0101, | |
| "eval_samples_per_second": 15.592, | |
| "eval_steps_per_second": 1.949, | |
| "eval_wer": 0.08168785501758183, | |
| "step": 1185 | |
| }, | |
| { | |
| "epoch": 15.19, | |
| "learning_rate": 2.5454430379746836e-05, | |
| "loss": 0.1695, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_cer": 0.02697425885012588, | |
| "eval_loss": 0.19071535766124725, | |
| "eval_runtime": 20.1621, | |
| "eval_samples_per_second": 15.475, | |
| "eval_steps_per_second": 1.934, | |
| "eval_wer": 0.08385177170678929, | |
| "step": 1264 | |
| }, | |
| { | |
| "epoch": 16.46, | |
| "learning_rate": 2.5074683544303798e-05, | |
| "loss": 0.1528, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_cer": 0.02856702461079998, | |
| "eval_loss": 0.18394820392131805, | |
| "eval_runtime": 19.8761, | |
| "eval_samples_per_second": 15.697, | |
| "eval_steps_per_second": 1.962, | |
| "eval_wer": 0.09061401136056262, | |
| "step": 1343 | |
| }, | |
| { | |
| "epoch": 17.72, | |
| "learning_rate": 2.469493670886076e-05, | |
| "loss": 0.1592, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_cer": 0.02810460874479782, | |
| "eval_loss": 0.18656496703624725, | |
| "eval_runtime": 20.0039, | |
| "eval_samples_per_second": 15.597, | |
| "eval_steps_per_second": 1.95, | |
| "eval_wer": 0.09034352177441168, | |
| "step": 1422 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 2.4315189873417722e-05, | |
| "loss": 0.1519, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_cer": 0.027488054256794944, | |
| "eval_loss": 0.20305851101875305, | |
| "eval_runtime": 20.3338, | |
| "eval_samples_per_second": 15.344, | |
| "eval_steps_per_second": 1.918, | |
| "eval_wer": 0.08574519880984582, | |
| "step": 1501 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.027796331500796384, | |
| "eval_loss": 0.19475489854812622, | |
| "eval_runtime": 19.9135, | |
| "eval_samples_per_second": 15.668, | |
| "eval_steps_per_second": 1.958, | |
| "eval_wer": 0.08601568839599676, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 20.25, | |
| "learning_rate": 2.3935443037974685e-05, | |
| "loss": 0.1257, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_cer": 0.026152186199455377, | |
| "eval_loss": 0.18495479226112366, | |
| "eval_runtime": 19.994, | |
| "eval_samples_per_second": 15.605, | |
| "eval_steps_per_second": 1.951, | |
| "eval_wer": 0.08601568839599676, | |
| "step": 1659 | |
| }, | |
| { | |
| "epoch": 21.52, | |
| "learning_rate": 2.3555696202531647e-05, | |
| "loss": 0.1288, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_cer": 0.02656322252479063, | |
| "eval_loss": 0.1773720234632492, | |
| "eval_runtime": 19.9911, | |
| "eval_samples_per_second": 15.607, | |
| "eval_steps_per_second": 1.951, | |
| "eval_wer": 0.08439275087909115, | |
| "step": 1738 | |
| }, | |
| { | |
| "epoch": 22.78, | |
| "learning_rate": 2.317594936708861e-05, | |
| "loss": 0.115, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.02651184298412372, | |
| "eval_loss": 0.19595499336719513, | |
| "eval_runtime": 19.9245, | |
| "eval_samples_per_second": 15.659, | |
| "eval_steps_per_second": 1.957, | |
| "eval_wer": 0.08439275087909115, | |
| "step": 1817 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_cer": 0.025843908955453937, | |
| "eval_loss": 0.18322429060935974, | |
| "eval_runtime": 19.9298, | |
| "eval_samples_per_second": 15.655, | |
| "eval_steps_per_second": 1.957, | |
| "eval_wer": 0.08249932377603462, | |
| "step": 1896 | |
| }, | |
| { | |
| "epoch": 24.05, | |
| "learning_rate": 2.2796202531645568e-05, | |
| "loss": 0.1223, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.02610080665878847, | |
| "eval_loss": 0.19195543229579926, | |
| "eval_runtime": 20.0272, | |
| "eval_samples_per_second": 15.579, | |
| "eval_steps_per_second": 1.947, | |
| "eval_wer": 0.08276981336218556, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 25.32, | |
| "learning_rate": 2.2416455696202534e-05, | |
| "loss": 0.1175, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_cer": 0.026049427118121565, | |
| "eval_loss": 0.1951374113559723, | |
| "eval_runtime": 20.0587, | |
| "eval_samples_per_second": 15.554, | |
| "eval_steps_per_second": 1.944, | |
| "eval_wer": 0.08033540708682715, | |
| "step": 2054 | |
| }, | |
| { | |
| "epoch": 26.58, | |
| "learning_rate": 2.2036708860759496e-05, | |
| "loss": 0.1051, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_cer": 0.02656322252479063, | |
| "eval_loss": 0.1995711475610733, | |
| "eval_runtime": 20.0348, | |
| "eval_samples_per_second": 15.573, | |
| "eval_steps_per_second": 1.947, | |
| "eval_wer": 0.08249932377603462, | |
| "step": 2133 | |
| }, | |
| { | |
| "epoch": 27.85, | |
| "learning_rate": 2.1656962025316455e-05, | |
| "loss": 0.1033, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_cer": 0.027385295175461132, | |
| "eval_loss": 0.21518437564373016, | |
| "eval_runtime": 20.185, | |
| "eval_samples_per_second": 15.457, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.08466324046524208, | |
| "step": 2212 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_cer": 0.02769357241946257, | |
| "eval_loss": 0.2082461565732956, | |
| "eval_runtime": 20.1321, | |
| "eval_samples_per_second": 15.498, | |
| "eval_steps_per_second": 1.937, | |
| "eval_wer": 0.08790911549905328, | |
| "step": 2291 | |
| }, | |
| { | |
| "epoch": 29.11, | |
| "learning_rate": 2.127721518987342e-05, | |
| "loss": 0.0961, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_cer": 0.027385295175461132, | |
| "eval_loss": 0.2153148353099823, | |
| "eval_runtime": 20.1045, | |
| "eval_samples_per_second": 15.519, | |
| "eval_steps_per_second": 1.94, | |
| "eval_wer": 0.08547470922369489, | |
| "step": 2370 | |
| }, | |
| { | |
| "epoch": 30.38, | |
| "learning_rate": 2.089746835443038e-05, | |
| "loss": 0.1003, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_cer": 0.028772542773467604, | |
| "eval_loss": 0.2043757140636444, | |
| "eval_runtime": 20.1868, | |
| "eval_samples_per_second": 15.456, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.09034352177441168, | |
| "step": 2449 | |
| }, | |
| { | |
| "epoch": 31.65, | |
| "learning_rate": 2.0517721518987342e-05, | |
| "loss": 0.1129, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_cer": 0.026768740687458253, | |
| "eval_loss": 0.20504404604434967, | |
| "eval_runtime": 20.2156, | |
| "eval_samples_per_second": 15.434, | |
| "eval_steps_per_second": 1.929, | |
| "eval_wer": 0.08547470922369489, | |
| "step": 2528 | |
| }, | |
| { | |
| "epoch": 32.91, | |
| "learning_rate": 2.0137974683544304e-05, | |
| "loss": 0.0939, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_cer": 0.0271283974721266, | |
| "eval_loss": 0.20279404520988464, | |
| "eval_runtime": 20.1612, | |
| "eval_samples_per_second": 15.475, | |
| "eval_steps_per_second": 1.934, | |
| "eval_wer": 0.08601568839599676, | |
| "step": 2607 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_cer": 0.027436674716128037, | |
| "eval_loss": 0.20313425362110138, | |
| "eval_runtime": 20.5421, | |
| "eval_samples_per_second": 15.188, | |
| "eval_steps_per_second": 1.899, | |
| "eval_wer": 0.08466324046524208, | |
| "step": 2686 | |
| }, | |
| { | |
| "epoch": 34.18, | |
| "learning_rate": 1.9758227848101266e-05, | |
| "loss": 0.0846, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_cer": 0.026922879309458973, | |
| "eval_loss": 0.20462077856063843, | |
| "eval_runtime": 20.1988, | |
| "eval_samples_per_second": 15.446, | |
| "eval_steps_per_second": 1.931, | |
| "eval_wer": 0.08222883418988369, | |
| "step": 2765 | |
| }, | |
| { | |
| "epoch": 35.44, | |
| "learning_rate": 1.937848101265823e-05, | |
| "loss": 0.083, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_cer": 0.02651184298412372, | |
| "eval_loss": 0.20938025414943695, | |
| "eval_runtime": 20.4858, | |
| "eval_samples_per_second": 15.23, | |
| "eval_steps_per_second": 1.904, | |
| "eval_wer": 0.08249932377603462, | |
| "step": 2844 | |
| }, | |
| { | |
| "epoch": 36.71, | |
| "learning_rate": 1.8998734177215188e-05, | |
| "loss": 0.0844, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_cer": 0.026768740687458253, | |
| "eval_loss": 0.21756267547607422, | |
| "eval_runtime": 20.1565, | |
| "eval_samples_per_second": 15.479, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.08195834460373276, | |
| "step": 2923 | |
| }, | |
| { | |
| "epoch": 37.97, | |
| "learning_rate": 1.8618987341772153e-05, | |
| "loss": 0.0829, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_cer": 0.02666598160612444, | |
| "eval_loss": 0.20818965137004852, | |
| "eval_runtime": 20.1541, | |
| "eval_samples_per_second": 15.481, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.08168785501758183, | |
| "step": 3002 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_cer": 0.028618404151466884, | |
| "eval_loss": 0.2200096845626831, | |
| "eval_runtime": 20.1838, | |
| "eval_samples_per_second": 15.458, | |
| "eval_steps_per_second": 1.932, | |
| "eval_wer": 0.08926156342980796, | |
| "step": 3081 | |
| }, | |
| { | |
| "epoch": 39.24, | |
| "learning_rate": 1.8239240506329115e-05, | |
| "loss": 0.103, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_cer": 0.027642192878795664, | |
| "eval_loss": 0.2102428525686264, | |
| "eval_runtime": 20.0863, | |
| "eval_samples_per_second": 15.533, | |
| "eval_steps_per_second": 1.942, | |
| "eval_wer": 0.08412226129294022, | |
| "step": 3160 | |
| }, | |
| { | |
| "epoch": 40.51, | |
| "learning_rate": 1.7859493670886074e-05, | |
| "loss": 0.0728, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_cer": 0.027077017931459692, | |
| "eval_loss": 0.2143275886774063, | |
| "eval_runtime": 20.2051, | |
| "eval_samples_per_second": 15.442, | |
| "eval_steps_per_second": 1.93, | |
| "eval_wer": 0.08168785501758183, | |
| "step": 3239 | |
| }, | |
| { | |
| "epoch": 41.77, | |
| "learning_rate": 1.747974683544304e-05, | |
| "loss": 0.079, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_cer": 0.02651184298412372, | |
| "eval_loss": 0.2131078988313675, | |
| "eval_runtime": 20.1578, | |
| "eval_samples_per_second": 15.478, | |
| "eval_steps_per_second": 1.935, | |
| "eval_wer": 0.08249932377603462, | |
| "step": 3318 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "step": 3318, | |
| "total_flos": 8.489873279367732e+19, | |
| "train_loss": 0.24954385395291484, | |
| "train_runtime": 14687.2421, | |
| "train_samples_per_second": 17.137, | |
| "train_steps_per_second": 0.538 | |
| } | |
| ], | |
| "max_steps": 7900, | |
| "num_train_epochs": 100, | |
| "total_flos": 8.489873279367732e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |