| { | |
| "best_metric": 0.294111043214798, | |
| "best_model_checkpoint": "wav2vec2-large-xlsr-mecita-coraa-portuguese-random-all-02/checkpoint-1978", | |
| "epoch": 43.0, | |
| "global_step": 3698, | |
| "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": 26.6237, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 0.7321738336528805, | |
| "eval_loss": 2.7489612102508545, | |
| "eval_runtime": 8.9523, | |
| "eval_samples_per_second": 38.091, | |
| "eval_steps_per_second": 4.803, | |
| "eval_wer": 0.9341623994147769, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "learning_rate": 2.966511627906977e-05, | |
| "loss": 5.3366, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 0.4202045101798594, | |
| "eval_loss": 1.6393556594848633, | |
| "eval_runtime": 8.8583, | |
| "eval_samples_per_second": 38.495, | |
| "eval_steps_per_second": 4.854, | |
| "eval_wer": 0.7008046817849305, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 2.33, | |
| "learning_rate": 2.9316279069767443e-05, | |
| "loss": 3.393, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 0.2307586962476034, | |
| "eval_loss": 1.015175700187683, | |
| "eval_runtime": 8.7977, | |
| "eval_samples_per_second": 38.76, | |
| "eval_steps_per_second": 4.888, | |
| "eval_wer": 0.49280663252865153, | |
| "step": 258 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 2.8967441860465116e-05, | |
| "loss": 2.1045, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 0.15105450561490003, | |
| "eval_loss": 0.6938430070877075, | |
| "eval_runtime": 8.4884, | |
| "eval_samples_per_second": 40.173, | |
| "eval_steps_per_second": 5.066, | |
| "eval_wer": 0.3511338697878566, | |
| "step": 344 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 2.861860465116279e-05, | |
| "loss": 1.8687, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_cer": 0.11325664201588606, | |
| "eval_loss": 0.5804391503334045, | |
| "eval_runtime": 8.8411, | |
| "eval_samples_per_second": 38.57, | |
| "eval_steps_per_second": 4.864, | |
| "eval_wer": 0.29748841745915633, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 5.81, | |
| "learning_rate": 2.8269767441860464e-05, | |
| "loss": 1.0808, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_cer": 0.09864877202592896, | |
| "eval_loss": 0.4938930869102478, | |
| "eval_runtime": 8.7485, | |
| "eval_samples_per_second": 38.978, | |
| "eval_steps_per_second": 4.915, | |
| "eval_wer": 0.2553035844915874, | |
| "step": 516 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "learning_rate": 2.7920930232558138e-05, | |
| "loss": 0.9453, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_cer": 0.08577558659727928, | |
| "eval_loss": 0.45988404750823975, | |
| "eval_runtime": 8.5437, | |
| "eval_samples_per_second": 39.912, | |
| "eval_steps_per_second": 5.033, | |
| "eval_wer": 0.23165081687393319, | |
| "step": 602 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_cer": 0.07390669223043915, | |
| "eval_loss": 0.4172210097312927, | |
| "eval_runtime": 8.6637, | |
| "eval_samples_per_second": 39.359, | |
| "eval_steps_per_second": 4.963, | |
| "eval_wer": 0.2138502804194099, | |
| "step": 688 | |
| }, | |
| { | |
| "epoch": 8.14, | |
| "learning_rate": 2.7572093023255812e-05, | |
| "loss": 0.7718, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_cer": 0.062174746644754864, | |
| "eval_loss": 0.37932026386260986, | |
| "eval_runtime": 8.5731, | |
| "eval_samples_per_second": 39.775, | |
| "eval_steps_per_second": 5.016, | |
| "eval_wer": 0.18068763716166789, | |
| "step": 774 | |
| }, | |
| { | |
| "epoch": 9.3, | |
| "learning_rate": 2.722325581395349e-05, | |
| "loss": 0.72, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_cer": 0.06358988405003195, | |
| "eval_loss": 0.4070007801055908, | |
| "eval_runtime": 8.5357, | |
| "eval_samples_per_second": 39.95, | |
| "eval_steps_per_second": 5.038, | |
| "eval_wer": 0.18141916605705927, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 10.47, | |
| "learning_rate": 2.6874418604651163e-05, | |
| "loss": 0.4718, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_cer": 0.056970692960832646, | |
| "eval_loss": 0.4255421757698059, | |
| "eval_runtime": 9.3648, | |
| "eval_samples_per_second": 36.413, | |
| "eval_steps_per_second": 4.592, | |
| "eval_wer": 0.16435015849792733, | |
| "step": 946 | |
| }, | |
| { | |
| "epoch": 11.63, | |
| "learning_rate": 2.6525581395348837e-05, | |
| "loss": 0.5514, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_cer": 0.0510818953711312, | |
| "eval_loss": 0.3528519868850708, | |
| "eval_runtime": 8.4978, | |
| "eval_samples_per_second": 40.128, | |
| "eval_steps_per_second": 5.06, | |
| "eval_wer": 0.15264569617166546, | |
| "step": 1032 | |
| }, | |
| { | |
| "epoch": 12.79, | |
| "learning_rate": 2.617674418604651e-05, | |
| "loss": 0.3473, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_cer": 0.05048845065278919, | |
| "eval_loss": 0.34577953815460205, | |
| "eval_runtime": 8.4874, | |
| "eval_samples_per_second": 40.177, | |
| "eval_steps_per_second": 5.066, | |
| "eval_wer": 0.1528895391367959, | |
| "step": 1118 | |
| }, | |
| { | |
| "epoch": 13.95, | |
| "learning_rate": 2.5827906976744185e-05, | |
| "loss": 0.4885, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_cer": 0.049575458778416875, | |
| "eval_loss": 0.3216852843761444, | |
| "eval_runtime": 8.6413, | |
| "eval_samples_per_second": 39.462, | |
| "eval_steps_per_second": 4.976, | |
| "eval_wer": 0.14435503535722993, | |
| "step": 1204 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_cer": 0.04802337259198393, | |
| "eval_loss": 0.31070542335510254, | |
| "eval_runtime": 8.5878, | |
| "eval_samples_per_second": 39.708, | |
| "eval_steps_per_second": 5.007, | |
| "eval_wer": 0.14484272128749084, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 2.5479069767441862e-05, | |
| "loss": 0.3774, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_cer": 0.0487994156852004, | |
| "eval_loss": 0.4008086621761322, | |
| "eval_runtime": 8.7031, | |
| "eval_samples_per_second": 39.181, | |
| "eval_steps_per_second": 4.941, | |
| "eval_wer": 0.14606193611314314, | |
| "step": 1376 | |
| }, | |
| { | |
| "epoch": 16.28, | |
| "learning_rate": 2.5130232558139536e-05, | |
| "loss": 0.3974, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_cer": 0.05012325390304026, | |
| "eval_loss": 0.38569143414497375, | |
| "eval_runtime": 8.6577, | |
| "eval_samples_per_second": 39.387, | |
| "eval_steps_per_second": 4.967, | |
| "eval_wer": 0.15191416727627408, | |
| "step": 1462 | |
| }, | |
| { | |
| "epoch": 17.44, | |
| "learning_rate": 2.478139534883721e-05, | |
| "loss": 0.3398, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_cer": 0.04222587418971971, | |
| "eval_loss": 0.3497083783149719, | |
| "eval_runtime": 8.7914, | |
| "eval_samples_per_second": 38.788, | |
| "eval_steps_per_second": 4.891, | |
| "eval_wer": 0.13094367227505485, | |
| "step": 1548 | |
| }, | |
| { | |
| "epoch": 18.6, | |
| "learning_rate": 2.4432558139534884e-05, | |
| "loss": 0.2741, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_cer": 0.042636720533187254, | |
| "eval_loss": 0.31500518321990967, | |
| "eval_runtime": 8.7036, | |
| "eval_samples_per_second": 39.179, | |
| "eval_steps_per_second": 4.94, | |
| "eval_wer": 0.13118751524018532, | |
| "step": 1634 | |
| }, | |
| { | |
| "epoch": 19.77, | |
| "learning_rate": 2.4083720930232558e-05, | |
| "loss": 0.2805, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_cer": 0.040491189628412305, | |
| "eval_loss": 0.3533484637737274, | |
| "eval_runtime": 8.6203, | |
| "eval_samples_per_second": 39.558, | |
| "eval_steps_per_second": 4.988, | |
| "eval_wer": 0.12362838332114119, | |
| "step": 1720 | |
| }, | |
| { | |
| "epoch": 20.93, | |
| "learning_rate": 2.373488372093023e-05, | |
| "loss": 0.3292, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_cer": 0.04240847256459417, | |
| "eval_loss": 0.32265007495880127, | |
| "eval_runtime": 8.657, | |
| "eval_samples_per_second": 39.39, | |
| "eval_steps_per_second": 4.967, | |
| "eval_wer": 0.12777371372835894, | |
| "step": 1806 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_cer": 0.04158677987765909, | |
| "eval_loss": 0.2969246506690979, | |
| "eval_runtime": 8.6619, | |
| "eval_samples_per_second": 39.368, | |
| "eval_steps_per_second": 4.964, | |
| "eval_wer": 0.12948061448427212, | |
| "step": 1892 | |
| }, | |
| { | |
| "epoch": 22.09, | |
| "learning_rate": 2.338604651162791e-05, | |
| "loss": 0.2255, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_cer": 0.03848260750479321, | |
| "eval_loss": 0.294111043214798, | |
| "eval_runtime": 8.643, | |
| "eval_samples_per_second": 39.454, | |
| "eval_steps_per_second": 4.975, | |
| "eval_wer": 0.11923920994879297, | |
| "step": 1978 | |
| }, | |
| { | |
| "epoch": 23.26, | |
| "learning_rate": 2.3037209302325583e-05, | |
| "loss": 0.2107, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_cer": 0.04213457500228248, | |
| "eval_loss": 0.3290224075317383, | |
| "eval_runtime": 8.3985, | |
| "eval_samples_per_second": 40.603, | |
| "eval_steps_per_second": 5.12, | |
| "eval_wer": 0.12606681297244574, | |
| "step": 2064 | |
| }, | |
| { | |
| "epoch": 24.42, | |
| "learning_rate": 2.2688372093023257e-05, | |
| "loss": 0.1922, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_cer": 0.039943394503788913, | |
| "eval_loss": 0.34917840361595154, | |
| "eval_runtime": 8.2683, | |
| "eval_samples_per_second": 41.242, | |
| "eval_steps_per_second": 5.201, | |
| "eval_wer": 0.12216532553035844, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 25.58, | |
| "learning_rate": 2.233953488372093e-05, | |
| "loss": 0.1829, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_cer": 0.03825435953620013, | |
| "eval_loss": 0.3639993667602539, | |
| "eval_runtime": 8.5021, | |
| "eval_samples_per_second": 40.108, | |
| "eval_steps_per_second": 5.058, | |
| "eval_wer": 0.11728846622774933, | |
| "step": 2236 | |
| }, | |
| { | |
| "epoch": 26.74, | |
| "learning_rate": 2.1990697674418605e-05, | |
| "loss": 0.1911, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_cer": 0.039578197754039986, | |
| "eval_loss": 0.3595152497291565, | |
| "eval_runtime": 8.5431, | |
| "eval_samples_per_second": 39.915, | |
| "eval_steps_per_second": 5.033, | |
| "eval_wer": 0.1224091684954889, | |
| "step": 2322 | |
| }, | |
| { | |
| "epoch": 27.91, | |
| "learning_rate": 2.1641860465116282e-05, | |
| "loss": 0.1712, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_cer": 0.03898475303569798, | |
| "eval_loss": 0.352108359336853, | |
| "eval_runtime": 8.4845, | |
| "eval_samples_per_second": 40.191, | |
| "eval_steps_per_second": 5.068, | |
| "eval_wer": 0.12192148256522799, | |
| "step": 2408 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_cer": 0.037432666849265044, | |
| "eval_loss": 0.3313020169734955, | |
| "eval_runtime": 8.6842, | |
| "eval_samples_per_second": 39.267, | |
| "eval_steps_per_second": 4.952, | |
| "eval_wer": 0.11363082175079249, | |
| "step": 2494 | |
| }, | |
| { | |
| "epoch": 29.07, | |
| "learning_rate": 2.1293023255813956e-05, | |
| "loss": 0.1708, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_cer": 0.03811741075504428, | |
| "eval_loss": 0.3219456076622009, | |
| "eval_runtime": 8.383, | |
| "eval_samples_per_second": 40.677, | |
| "eval_steps_per_second": 5.129, | |
| "eval_wer": 0.12070226773957571, | |
| "step": 2580 | |
| }, | |
| { | |
| "epoch": 30.23, | |
| "learning_rate": 2.094418604651163e-05, | |
| "loss": 0.1389, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_cer": 0.03610882863142518, | |
| "eval_loss": 0.3261343240737915, | |
| "eval_runtime": 8.5295, | |
| "eval_samples_per_second": 39.979, | |
| "eval_steps_per_second": 5.041, | |
| "eval_wer": 0.11143623506461839, | |
| "step": 2666 | |
| }, | |
| { | |
| "epoch": 31.4, | |
| "learning_rate": 2.0595348837209304e-05, | |
| "loss": 0.1516, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_cer": 0.035926230256550715, | |
| "eval_loss": 0.3445923328399658, | |
| "eval_runtime": 8.7722, | |
| "eval_samples_per_second": 38.873, | |
| "eval_steps_per_second": 4.902, | |
| "eval_wer": 0.11021702023896611, | |
| "step": 2752 | |
| }, | |
| { | |
| "epoch": 32.56, | |
| "learning_rate": 2.025e-05, | |
| "loss": 0.2601, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_cer": 0.03670227334976719, | |
| "eval_loss": 0.35051313042640686, | |
| "eval_runtime": 8.461, | |
| "eval_samples_per_second": 40.303, | |
| "eval_steps_per_second": 5.082, | |
| "eval_wer": 0.11509387954157523, | |
| "step": 2838 | |
| }, | |
| { | |
| "epoch": 33.72, | |
| "learning_rate": 1.9901162790697675e-05, | |
| "loss": 0.1392, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_cer": 0.03670227334976719, | |
| "eval_loss": 0.3281579315662384, | |
| "eval_runtime": 8.5326, | |
| "eval_samples_per_second": 39.964, | |
| "eval_steps_per_second": 5.039, | |
| "eval_wer": 0.11314313582053158, | |
| "step": 2924 | |
| }, | |
| { | |
| "epoch": 34.88, | |
| "learning_rate": 1.955232558139535e-05, | |
| "loss": 0.1286, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_cer": 0.035926230256550715, | |
| "eval_loss": 0.3351230025291443, | |
| "eval_runtime": 9.1156, | |
| "eval_samples_per_second": 37.408, | |
| "eval_steps_per_second": 4.717, | |
| "eval_wer": 0.11289929285540112, | |
| "step": 3010 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_cer": 0.03565233269423902, | |
| "eval_loss": 0.348171591758728, | |
| "eval_runtime": 8.8289, | |
| "eval_samples_per_second": 38.623, | |
| "eval_steps_per_second": 4.87, | |
| "eval_wer": 0.1119239209948793, | |
| "step": 3096 | |
| }, | |
| { | |
| "epoch": 36.05, | |
| "learning_rate": 1.9203488372093023e-05, | |
| "loss": 0.1497, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_cer": 0.03734136766182781, | |
| "eval_loss": 0.37620189785957336, | |
| "eval_runtime": 8.8804, | |
| "eval_samples_per_second": 38.399, | |
| "eval_steps_per_second": 4.842, | |
| "eval_wer": 0.11558156547183614, | |
| "step": 3182 | |
| }, | |
| { | |
| "epoch": 37.21, | |
| "learning_rate": 1.8854651162790697e-05, | |
| "loss": 0.1319, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_cer": 0.03725006847439058, | |
| "eval_loss": 0.3732824921607971, | |
| "eval_runtime": 8.3326, | |
| "eval_samples_per_second": 40.924, | |
| "eval_steps_per_second": 5.16, | |
| "eval_wer": 0.11485003657644477, | |
| "step": 3268 | |
| }, | |
| { | |
| "epoch": 38.37, | |
| "learning_rate": 1.8505813953488374e-05, | |
| "loss": 0.1294, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_cer": 0.037158769286953344, | |
| "eval_loss": 0.3463388681411743, | |
| "eval_runtime": 8.7462, | |
| "eval_samples_per_second": 38.988, | |
| "eval_steps_per_second": 4.916, | |
| "eval_wer": 0.11777615215801024, | |
| "step": 3354 | |
| }, | |
| { | |
| "epoch": 39.53, | |
| "learning_rate": 1.8156976744186048e-05, | |
| "loss": 0.1459, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_cer": 0.03784351319273258, | |
| "eval_loss": 0.34395572543144226, | |
| "eval_runtime": 8.9131, | |
| "eval_samples_per_second": 38.258, | |
| "eval_steps_per_second": 4.824, | |
| "eval_wer": 0.11460619361131431, | |
| "step": 3440 | |
| }, | |
| { | |
| "epoch": 40.7, | |
| "learning_rate": 1.780813953488372e-05, | |
| "loss": 0.0998, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_cer": 0.03656532456861134, | |
| "eval_loss": 0.34668025374412537, | |
| "eval_runtime": 8.643, | |
| "eval_samples_per_second": 39.454, | |
| "eval_steps_per_second": 4.975, | |
| "eval_wer": 0.11314313582053158, | |
| "step": 3526 | |
| }, | |
| { | |
| "epoch": 41.86, | |
| "learning_rate": 1.7459302325581396e-05, | |
| "loss": 0.1036, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_cer": 0.036200127818862414, | |
| "eval_loss": 0.32251447439193726, | |
| "eval_runtime": 8.672, | |
| "eval_samples_per_second": 39.322, | |
| "eval_steps_per_second": 4.959, | |
| "eval_wer": 0.11265544989027067, | |
| "step": 3612 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_cer": 0.03505888797589701, | |
| "eval_loss": 0.3630313277244568, | |
| "eval_runtime": 8.7676, | |
| "eval_samples_per_second": 38.893, | |
| "eval_steps_per_second": 4.904, | |
| "eval_wer": 0.11046086320409657, | |
| "step": 3698 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "step": 3698, | |
| "total_flos": 9.439881750406878e+19, | |
| "train_loss": 0.6348635904075908, | |
| "train_runtime": 13996.2801, | |
| "train_samples_per_second": 19.598, | |
| "train_steps_per_second": 0.614 | |
| } | |
| ], | |
| "max_steps": 8600, | |
| "num_train_epochs": 100, | |
| "total_flos": 9.439881750406878e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |