| { | |
| "best_metric": 0.6787810921669006, | |
| "best_model_checkpoint": "./wav2vec2-large-960h-lv60-self-paper/checkpoint-4190", | |
| "epoch": 50.0, | |
| "global_step": 20950, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 3.3473238945007324, | |
| "eval_runtime": 55.1245, | |
| "eval_samples_per_second": 15.275, | |
| "eval_steps_per_second": 1.923, | |
| "eval_wer": 1.0, | |
| "step": 419 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.981977593765222e-05, | |
| "loss": 5.8068, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.9190704822540283, | |
| "eval_runtime": 55.4984, | |
| "eval_samples_per_second": 15.172, | |
| "eval_steps_per_second": 1.91, | |
| "eval_wer": 0.8916934741113499, | |
| "step": 838 | |
| }, | |
| { | |
| "epoch": 2.39, | |
| "learning_rate": 4.860204578665368e-05, | |
| "loss": 2.5663, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 1.100569248199463, | |
| "eval_runtime": 55.6905, | |
| "eval_samples_per_second": 15.119, | |
| "eval_steps_per_second": 1.903, | |
| "eval_wer": 0.5802154384185165, | |
| "step": 1257 | |
| }, | |
| { | |
| "epoch": 3.58, | |
| "learning_rate": 4.7384315635655144e-05, | |
| "loss": 1.1433, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.9009209275245667, | |
| "eval_runtime": 55.6736, | |
| "eval_samples_per_second": 15.124, | |
| "eval_steps_per_second": 1.904, | |
| "eval_wer": 0.48139941535299063, | |
| "step": 1676 | |
| }, | |
| { | |
| "epoch": 4.77, | |
| "learning_rate": 4.6166585484656604e-05, | |
| "loss": 0.8522, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.8214807510375977, | |
| "eval_runtime": 55.5868, | |
| "eval_samples_per_second": 15.147, | |
| "eval_steps_per_second": 1.907, | |
| "eval_wer": 0.42471001241373785, | |
| "step": 2095 | |
| }, | |
| { | |
| "epoch": 5.97, | |
| "learning_rate": 4.4948855333658065e-05, | |
| "loss": 0.7256, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.7522485256195068, | |
| "eval_runtime": 55.6074, | |
| "eval_samples_per_second": 15.142, | |
| "eval_steps_per_second": 1.906, | |
| "eval_wer": 0.39215398374200783, | |
| "step": 2514 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.7201786041259766, | |
| "eval_runtime": 55.4815, | |
| "eval_samples_per_second": 15.176, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.3653643365324292, | |
| "step": 2933 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 4.3731125182659525e-05, | |
| "loss": 0.6239, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.6909050941467285, | |
| "eval_runtime": 55.6124, | |
| "eval_samples_per_second": 15.14, | |
| "eval_steps_per_second": 1.906, | |
| "eval_wer": 0.3579427900209565, | |
| "step": 3352 | |
| }, | |
| { | |
| "epoch": 8.35, | |
| "learning_rate": 4.2513395031660985e-05, | |
| "loss": 0.5618, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.6887015700340271, | |
| "eval_runtime": 56.9486, | |
| "eval_samples_per_second": 14.785, | |
| "eval_steps_per_second": 1.861, | |
| "eval_wer": 0.340002936583152, | |
| "step": 3771 | |
| }, | |
| { | |
| "epoch": 9.55, | |
| "learning_rate": 4.129566488066245e-05, | |
| "loss": 0.4998, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.6787810921669006, | |
| "eval_runtime": 57.0323, | |
| "eval_samples_per_second": 14.764, | |
| "eval_steps_per_second": 1.859, | |
| "eval_wer": 0.3320341177569844, | |
| "step": 4190 | |
| }, | |
| { | |
| "epoch": 10.74, | |
| "learning_rate": 4.0077934729663906e-05, | |
| "loss": 0.4569, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.680539071559906, | |
| "eval_runtime": 55.5776, | |
| "eval_samples_per_second": 15.15, | |
| "eval_steps_per_second": 1.907, | |
| "eval_wer": 0.3351308781718435, | |
| "step": 4609 | |
| }, | |
| { | |
| "epoch": 11.93, | |
| "learning_rate": 3.886020457866537e-05, | |
| "loss": 0.4156, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.6909714937210083, | |
| "eval_runtime": 55.4928, | |
| "eval_samples_per_second": 15.173, | |
| "eval_steps_per_second": 1.91, | |
| "eval_wer": 0.3253066727178077, | |
| "step": 5028 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_loss": 0.6859297752380371, | |
| "eval_runtime": 55.5121, | |
| "eval_samples_per_second": 15.168, | |
| "eval_steps_per_second": 1.909, | |
| "eval_wer": 0.32790955323891774, | |
| "step": 5447 | |
| }, | |
| { | |
| "epoch": 13.13, | |
| "learning_rate": 3.7642474427666826e-05, | |
| "loss": 0.3763, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 0.7075064778327942, | |
| "eval_runtime": 56.9008, | |
| "eval_samples_per_second": 14.798, | |
| "eval_steps_per_second": 1.863, | |
| "eval_wer": 0.32066153209551906, | |
| "step": 5866 | |
| }, | |
| { | |
| "epoch": 14.32, | |
| "learning_rate": 3.6424744276668293e-05, | |
| "loss": 0.3473, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_loss": 0.7173970341682434, | |
| "eval_runtime": 56.8103, | |
| "eval_samples_per_second": 14.821, | |
| "eval_steps_per_second": 1.866, | |
| "eval_wer": 0.3151754608433333, | |
| "step": 6285 | |
| }, | |
| { | |
| "epoch": 15.51, | |
| "learning_rate": 3.5207014125669754e-05, | |
| "loss": 0.3141, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.7283802628517151, | |
| "eval_runtime": 55.482, | |
| "eval_samples_per_second": 15.176, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.31707089178691084, | |
| "step": 6704 | |
| }, | |
| { | |
| "epoch": 16.71, | |
| "learning_rate": 3.3989283974671214e-05, | |
| "loss": 0.2884, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_loss": 0.7537466883659363, | |
| "eval_runtime": 55.5565, | |
| "eval_samples_per_second": 15.156, | |
| "eval_steps_per_second": 1.908, | |
| "eval_wer": 0.3192332848352176, | |
| "step": 7123 | |
| }, | |
| { | |
| "epoch": 17.9, | |
| "learning_rate": 3.277398928397467e-05, | |
| "loss": 0.2771, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_loss": 0.7311689853668213, | |
| "eval_runtime": 55.4309, | |
| "eval_samples_per_second": 15.19, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.31748468304924116, | |
| "step": 7542 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_loss": 0.7669196128845215, | |
| "eval_runtime": 55.6417, | |
| "eval_samples_per_second": 15.133, | |
| "eval_steps_per_second": 1.905, | |
| "eval_wer": 0.3138139541092142, | |
| "step": 7961 | |
| }, | |
| { | |
| "epoch": 19.09, | |
| "learning_rate": 3.155625913297613e-05, | |
| "loss": 0.2538, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_loss": 0.8143336772918701, | |
| "eval_runtime": 55.4405, | |
| "eval_samples_per_second": 15.187, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.30738016738523966, | |
| "step": 8380 | |
| }, | |
| { | |
| "epoch": 20.29, | |
| "learning_rate": 3.0338528981977592e-05, | |
| "loss": 0.2319, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_loss": 0.8184694647789001, | |
| "eval_runtime": 55.5569, | |
| "eval_samples_per_second": 15.156, | |
| "eval_steps_per_second": 1.908, | |
| "eval_wer": 0.30880841464554104, | |
| "step": 8799 | |
| }, | |
| { | |
| "epoch": 21.48, | |
| "learning_rate": 2.9120798830979056e-05, | |
| "loss": 0.2206, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.8111276626586914, | |
| "eval_runtime": 55.431, | |
| "eval_samples_per_second": 15.19, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.30689963559672706, | |
| "step": 9218 | |
| }, | |
| { | |
| "epoch": 22.67, | |
| "learning_rate": 2.7903068679980516e-05, | |
| "loss": 0.2093, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_loss": 0.8248062133789062, | |
| "eval_runtime": 55.4548, | |
| "eval_samples_per_second": 15.184, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.30880841464554104, | |
| "step": 9637 | |
| }, | |
| { | |
| "epoch": 23.87, | |
| "learning_rate": 2.668533852898198e-05, | |
| "loss": 0.1979, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_loss": 0.8571637868881226, | |
| "eval_runtime": 55.4105, | |
| "eval_samples_per_second": 15.196, | |
| "eval_steps_per_second": 1.913, | |
| "eval_wer": 0.3067127621234166, | |
| "step": 10056 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_loss": 0.8709866404533386, | |
| "eval_runtime": 55.5254, | |
| "eval_samples_per_second": 15.164, | |
| "eval_steps_per_second": 1.909, | |
| "eval_wer": 0.3073935154904761, | |
| "step": 10475 | |
| }, | |
| { | |
| "epoch": 25.06, | |
| "learning_rate": 2.5467608377983437e-05, | |
| "loss": 0.1852, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_loss": 0.8921730518341064, | |
| "eval_runtime": 55.3415, | |
| "eval_samples_per_second": 15.215, | |
| "eval_steps_per_second": 1.915, | |
| "eval_wer": 0.3067394583338895, | |
| "step": 10894 | |
| }, | |
| { | |
| "epoch": 26.25, | |
| "learning_rate": 2.42498782269849e-05, | |
| "loss": 0.1742, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_loss": 0.9039767384529114, | |
| "eval_runtime": 55.3805, | |
| "eval_samples_per_second": 15.204, | |
| "eval_steps_per_second": 1.914, | |
| "eval_wer": 0.30679285075483537, | |
| "step": 11313 | |
| }, | |
| { | |
| "epoch": 27.45, | |
| "learning_rate": 2.303214807598636e-05, | |
| "loss": 0.1688, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_loss": 0.9143679738044739, | |
| "eval_runtime": 55.4828, | |
| "eval_samples_per_second": 15.176, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.30160043781785173, | |
| "step": 11732 | |
| }, | |
| { | |
| "epoch": 28.64, | |
| "learning_rate": 2.1816853385289822e-05, | |
| "loss": 0.1578, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_loss": 0.8989725112915039, | |
| "eval_runtime": 55.3269, | |
| "eval_samples_per_second": 15.219, | |
| "eval_steps_per_second": 1.916, | |
| "eval_wer": 0.310917415272902, | |
| "step": 12151 | |
| }, | |
| { | |
| "epoch": 29.83, | |
| "learning_rate": 2.0599123234291282e-05, | |
| "loss": 0.1557, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_loss": 0.9465027451515198, | |
| "eval_runtime": 55.5076, | |
| "eval_samples_per_second": 15.169, | |
| "eval_steps_per_second": 1.91, | |
| "eval_wer": 0.30042580455704315, | |
| "step": 12570 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_loss": 0.9480117559432983, | |
| "eval_runtime": 56.1121, | |
| "eval_samples_per_second": 15.006, | |
| "eval_steps_per_second": 1.889, | |
| "eval_wer": 0.3024680646582218, | |
| "step": 12989 | |
| }, | |
| { | |
| "epoch": 31.03, | |
| "learning_rate": 1.9381393083292742e-05, | |
| "loss": 0.1456, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_loss": 0.9730611443519592, | |
| "eval_runtime": 55.4397, | |
| "eval_samples_per_second": 15.188, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.3016538302387976, | |
| "step": 13408 | |
| }, | |
| { | |
| "epoch": 32.22, | |
| "learning_rate": 1.81660983925962e-05, | |
| "loss": 0.1398, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_loss": 0.9633293151855469, | |
| "eval_runtime": 55.4286, | |
| "eval_samples_per_second": 15.191, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.30380287518186794, | |
| "step": 13827 | |
| }, | |
| { | |
| "epoch": 33.41, | |
| "learning_rate": 1.6948368241597664e-05, | |
| "loss": 0.1343, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_loss": 0.9843713045120239, | |
| "eval_runtime": 55.4228, | |
| "eval_samples_per_second": 15.192, | |
| "eval_steps_per_second": 1.913, | |
| "eval_wer": 0.30114660223981204, | |
| "step": 14246 | |
| }, | |
| { | |
| "epoch": 34.61, | |
| "learning_rate": 1.5730638090599124e-05, | |
| "loss": 0.1275, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_loss": 1.007832407951355, | |
| "eval_runtime": 55.4358, | |
| "eval_samples_per_second": 15.189, | |
| "eval_steps_per_second": 1.912, | |
| "eval_wer": 0.2996783106638013, | |
| "step": 14665 | |
| }, | |
| { | |
| "epoch": 35.8, | |
| "learning_rate": 1.4512907939600584e-05, | |
| "loss": 0.1266, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_loss": 1.0066460371017456, | |
| "eval_runtime": 55.4803, | |
| "eval_samples_per_second": 15.177, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.2996382663480919, | |
| "step": 15084 | |
| }, | |
| { | |
| "epoch": 36.99, | |
| "learning_rate": 1.3295177788602044e-05, | |
| "loss": 0.1243, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_loss": 1.0132853984832764, | |
| "eval_runtime": 55.3704, | |
| "eval_samples_per_second": 15.207, | |
| "eval_steps_per_second": 1.914, | |
| "eval_wer": 0.3014135643445413, | |
| "step": 15503 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_loss": 1.0387077331542969, | |
| "eval_runtime": 55.3084, | |
| "eval_samples_per_second": 15.224, | |
| "eval_steps_per_second": 1.917, | |
| "eval_wer": 0.2971822149845829, | |
| "step": 15922 | |
| }, | |
| { | |
| "epoch": 38.19, | |
| "learning_rate": 1.2077447637603508e-05, | |
| "loss": 0.1182, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_loss": 1.0173468589782715, | |
| "eval_runtime": 55.2898, | |
| "eval_samples_per_second": 15.229, | |
| "eval_steps_per_second": 1.917, | |
| "eval_wer": 0.3026015457105864, | |
| "step": 16341 | |
| }, | |
| { | |
| "epoch": 39.38, | |
| "learning_rate": 1.0859717486604968e-05, | |
| "loss": 0.1152, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_loss": 1.0526808500289917, | |
| "eval_runtime": 55.2419, | |
| "eval_samples_per_second": 15.242, | |
| "eval_steps_per_second": 1.919, | |
| "eval_wer": 0.2977161391940414, | |
| "step": 16760 | |
| }, | |
| { | |
| "epoch": 40.57, | |
| "learning_rate": 9.644422795908426e-06, | |
| "loss": 0.1134, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_loss": 1.0490872859954834, | |
| "eval_runtime": 56.6858, | |
| "eval_samples_per_second": 14.854, | |
| "eval_steps_per_second": 1.87, | |
| "eval_wer": 0.29776953161498726, | |
| "step": 17179 | |
| }, | |
| { | |
| "epoch": 41.77, | |
| "learning_rate": 8.426692644909888e-06, | |
| "loss": 0.1101, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_loss": 1.0661942958831787, | |
| "eval_runtime": 56.702, | |
| "eval_samples_per_second": 14.85, | |
| "eval_steps_per_second": 1.869, | |
| "eval_wer": 0.2975693100364403, | |
| "step": 17598 | |
| }, | |
| { | |
| "epoch": 42.96, | |
| "learning_rate": 7.208962493911349e-06, | |
| "loss": 0.1083, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_loss": 1.0543538331985474, | |
| "eval_runtime": 55.2662, | |
| "eval_samples_per_second": 15.235, | |
| "eval_steps_per_second": 1.918, | |
| "eval_wer": 0.29790301266735186, | |
| "step": 18017 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_loss": 1.0599024295806885, | |
| "eval_runtime": 55.6295, | |
| "eval_samples_per_second": 15.136, | |
| "eval_steps_per_second": 1.905, | |
| "eval_wer": 0.29568722719809926, | |
| "step": 18436 | |
| }, | |
| { | |
| "epoch": 44.15, | |
| "learning_rate": 5.991232342912811e-06, | |
| "loss": 0.1073, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_loss": 1.0766526460647583, | |
| "eval_runtime": 55.4679, | |
| "eval_samples_per_second": 15.18, | |
| "eval_steps_per_second": 1.911, | |
| "eval_wer": 0.2959141449871191, | |
| "step": 18855 | |
| }, | |
| { | |
| "epoch": 45.35, | |
| "learning_rate": 4.773502191914271e-06, | |
| "loss": 0.1045, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_loss": 1.0773001909255981, | |
| "eval_runtime": 55.3004, | |
| "eval_samples_per_second": 15.226, | |
| "eval_steps_per_second": 1.917, | |
| "eval_wer": 0.29592749309235555, | |
| "step": 19274 | |
| }, | |
| { | |
| "epoch": 46.54, | |
| "learning_rate": 3.5557720409157334e-06, | |
| "loss": 0.1024, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_loss": 1.0730717182159424, | |
| "eval_runtime": 55.5009, | |
| "eval_samples_per_second": 15.171, | |
| "eval_steps_per_second": 1.91, | |
| "eval_wer": 0.2952600878305325, | |
| "step": 19693 | |
| }, | |
| { | |
| "epoch": 47.73, | |
| "learning_rate": 2.3380418899171945e-06, | |
| "loss": 0.1015, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_loss": 1.0823478698730469, | |
| "eval_runtime": 55.423, | |
| "eval_samples_per_second": 15.192, | |
| "eval_steps_per_second": 1.913, | |
| "eval_wer": 0.2965682021437057, | |
| "step": 20112 | |
| }, | |
| { | |
| "epoch": 48.93, | |
| "learning_rate": 1.1203117389186556e-06, | |
| "loss": 0.1016, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_loss": 1.08854079246521, | |
| "eval_runtime": 55.6127, | |
| "eval_samples_per_second": 15.14, | |
| "eval_steps_per_second": 1.906, | |
| "eval_wer": 0.29445920151634475, | |
| "step": 20531 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_loss": 1.0854156017303467, | |
| "eval_runtime": 55.4189, | |
| "eval_samples_per_second": 15.193, | |
| "eval_steps_per_second": 1.913, | |
| "eval_wer": 0.2950331700415126, | |
| "step": 20950 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "step": 20950, | |
| "total_flos": 1.40587009875955e+20, | |
| "train_loss": 0.45776011664998273, | |
| "train_runtime": 43219.7022, | |
| "train_samples_per_second": 3.876, | |
| "train_steps_per_second": 0.485 | |
| } | |
| ], | |
| "max_steps": 20950, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.40587009875955e+20, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |