| { | |
| "best_metric": 0.10026070475578308, | |
| "best_model_checkpoint": "./vit-base-ecg/checkpoint-300", | |
| "epoch": 20.0, | |
| "eval_steps": 100, | |
| "global_step": 820, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.24390243902439024, | |
| "grad_norm": 3.298177480697632, | |
| "learning_rate": 0.0001975609756097561, | |
| "loss": 1.3469, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.4878048780487805, | |
| "grad_norm": 1.9206117391586304, | |
| "learning_rate": 0.0001951219512195122, | |
| "loss": 1.0603, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.7317073170731707, | |
| "grad_norm": 3.621140956878662, | |
| "learning_rate": 0.0001926829268292683, | |
| "loss": 0.9322, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.975609756097561, | |
| "grad_norm": 7.4158735275268555, | |
| "learning_rate": 0.0001902439024390244, | |
| "loss": 0.7773, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.2195121951219512, | |
| "grad_norm": 4.021040916442871, | |
| "learning_rate": 0.0001878048780487805, | |
| "loss": 0.5297, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4634146341463414, | |
| "grad_norm": 2.9927070140838623, | |
| "learning_rate": 0.0001853658536585366, | |
| "loss": 0.4686, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.7073170731707317, | |
| "grad_norm": 1.2623863220214844, | |
| "learning_rate": 0.0001829268292682927, | |
| "loss": 0.4439, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.951219512195122, | |
| "grad_norm": 10.744318962097168, | |
| "learning_rate": 0.0001804878048780488, | |
| "loss": 0.3923, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.1951219512195124, | |
| "grad_norm": 8.121638298034668, | |
| "learning_rate": 0.00017804878048780488, | |
| "loss": 0.2701, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "grad_norm": 6.202785015106201, | |
| "learning_rate": 0.000175609756097561, | |
| "loss": 0.596, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.4390243902439024, | |
| "eval_accuracy": 0.8214285714285714, | |
| "eval_loss": 0.5431132912635803, | |
| "eval_runtime": 8.502, | |
| "eval_samples_per_second": 16.467, | |
| "eval_steps_per_second": 2.117, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.682926829268293, | |
| "grad_norm": 1.8862333297729492, | |
| "learning_rate": 0.00017317073170731708, | |
| "loss": 0.3225, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.926829268292683, | |
| "grad_norm": 2.015392780303955, | |
| "learning_rate": 0.0001707317073170732, | |
| "loss": 0.3347, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.1707317073170733, | |
| "grad_norm": 1.7934260368347168, | |
| "learning_rate": 0.00016829268292682927, | |
| "loss": 0.202, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.4146341463414633, | |
| "grad_norm": 2.0693511962890625, | |
| "learning_rate": 0.00016585365853658536, | |
| "loss": 0.2906, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.658536585365854, | |
| "grad_norm": 3.4595236778259277, | |
| "learning_rate": 0.00016341463414634147, | |
| "loss": 0.2004, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.902439024390244, | |
| "grad_norm": 3.176142692565918, | |
| "learning_rate": 0.00016097560975609758, | |
| "loss": 0.1719, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.146341463414634, | |
| "grad_norm": 1.0586614608764648, | |
| "learning_rate": 0.00015853658536585366, | |
| "loss": 0.0913, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.390243902439025, | |
| "grad_norm": 0.48713743686676025, | |
| "learning_rate": 0.00015609756097560978, | |
| "loss": 0.0923, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.634146341463414, | |
| "grad_norm": 0.5843382477760315, | |
| "learning_rate": 0.00015365853658536586, | |
| "loss": 0.1999, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "grad_norm": 0.11404013633728027, | |
| "learning_rate": 0.00015121951219512197, | |
| "loss": 0.0656, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.878048780487805, | |
| "eval_accuracy": 0.95, | |
| "eval_loss": 0.16283166408538818, | |
| "eval_runtime": 8.0607, | |
| "eval_samples_per_second": 17.368, | |
| "eval_steps_per_second": 2.233, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.121951219512195, | |
| "grad_norm": 0.10728397220373154, | |
| "learning_rate": 0.00014878048780487806, | |
| "loss": 0.0267, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.365853658536586, | |
| "grad_norm": 3.2944324016571045, | |
| "learning_rate": 0.00014634146341463414, | |
| "loss": 0.1612, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.609756097560975, | |
| "grad_norm": 0.12557579576969147, | |
| "learning_rate": 0.00014390243902439025, | |
| "loss": 0.1216, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.853658536585366, | |
| "grad_norm": 6.2914252281188965, | |
| "learning_rate": 0.00014146341463414634, | |
| "loss": 0.1067, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.097560975609756, | |
| "grad_norm": 3.4353785514831543, | |
| "learning_rate": 0.00013902439024390245, | |
| "loss": 0.0795, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.341463414634147, | |
| "grad_norm": 0.08993230760097504, | |
| "learning_rate": 0.00013658536585365856, | |
| "loss": 0.0919, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.585365853658536, | |
| "grad_norm": 7.937359809875488, | |
| "learning_rate": 0.00013414634146341464, | |
| "loss": 0.0621, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.829268292682927, | |
| "grad_norm": 0.10531166195869446, | |
| "learning_rate": 0.00013170731707317076, | |
| "loss": 0.0341, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.073170731707317, | |
| "grad_norm": 0.33791643381118774, | |
| "learning_rate": 0.00012926829268292684, | |
| "loss": 0.047, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "grad_norm": 0.06028667464852333, | |
| "learning_rate": 0.00012682926829268293, | |
| "loss": 0.0192, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.317073170731708, | |
| "eval_accuracy": 0.9642857142857143, | |
| "eval_loss": 0.10026070475578308, | |
| "eval_runtime": 7.8254, | |
| "eval_samples_per_second": 17.89, | |
| "eval_steps_per_second": 2.3, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.560975609756097, | |
| "grad_norm": 0.11959153413772583, | |
| "learning_rate": 0.00012439024390243904, | |
| "loss": 0.0145, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.804878048780488, | |
| "grad_norm": 0.05471642687916756, | |
| "learning_rate": 0.00012195121951219512, | |
| "loss": 0.0987, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.048780487804878, | |
| "grad_norm": 0.0556318461894989, | |
| "learning_rate": 0.00011951219512195122, | |
| "loss": 0.0447, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.292682926829269, | |
| "grad_norm": 0.04877757653594017, | |
| "learning_rate": 0.00011707317073170732, | |
| "loss": 0.0226, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.536585365853659, | |
| "grad_norm": 0.6035917401313782, | |
| "learning_rate": 0.00011463414634146342, | |
| "loss": 0.0465, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.78048780487805, | |
| "grad_norm": 0.04456908255815506, | |
| "learning_rate": 0.00011219512195121953, | |
| "loss": 0.0116, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.024390243902438, | |
| "grad_norm": 0.04949244484305382, | |
| "learning_rate": 0.00010975609756097563, | |
| "loss": 0.0349, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.268292682926829, | |
| "grad_norm": 0.047532856464385986, | |
| "learning_rate": 0.00010731707317073172, | |
| "loss": 0.0097, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.512195121951219, | |
| "grad_norm": 7.109127521514893, | |
| "learning_rate": 0.00010487804878048781, | |
| "loss": 0.0565, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "grad_norm": 0.04661267623305321, | |
| "learning_rate": 0.0001024390243902439, | |
| "loss": 0.0926, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.75609756097561, | |
| "eval_accuracy": 0.95, | |
| "eval_loss": 0.1261894255876541, | |
| "eval_runtime": 7.971, | |
| "eval_samples_per_second": 17.564, | |
| "eval_steps_per_second": 2.258, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.12421295046806335, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0375, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.24390243902439, | |
| "grad_norm": 0.04042995348572731, | |
| "learning_rate": 9.75609756097561e-05, | |
| "loss": 0.0199, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.487804878048781, | |
| "grad_norm": 0.05262177810072899, | |
| "learning_rate": 9.51219512195122e-05, | |
| "loss": 0.0095, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.731707317073171, | |
| "grad_norm": 0.03628499433398247, | |
| "learning_rate": 9.26829268292683e-05, | |
| "loss": 0.0078, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.975609756097562, | |
| "grad_norm": 0.04206101596355438, | |
| "learning_rate": 9.02439024390244e-05, | |
| "loss": 0.0369, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.21951219512195, | |
| "grad_norm": 0.03796745836734772, | |
| "learning_rate": 8.78048780487805e-05, | |
| "loss": 0.0081, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.463414634146341, | |
| "grad_norm": 0.03188019618391991, | |
| "learning_rate": 8.53658536585366e-05, | |
| "loss": 0.0075, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 11.707317073170731, | |
| "grad_norm": 0.04463886842131615, | |
| "learning_rate": 8.292682926829268e-05, | |
| "loss": 0.0071, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 11.951219512195122, | |
| "grad_norm": 0.024716157466173172, | |
| "learning_rate": 8.048780487804879e-05, | |
| "loss": 0.0066, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.195121951219512, | |
| "grad_norm": 0.030284544453024864, | |
| "learning_rate": 7.804878048780489e-05, | |
| "loss": 0.0064, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.195121951219512, | |
| "eval_accuracy": 0.9642857142857143, | |
| "eval_loss": 0.1611199826002121, | |
| "eval_runtime": 8.7409, | |
| "eval_samples_per_second": 16.017, | |
| "eval_steps_per_second": 2.059, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 12.439024390243903, | |
| "grad_norm": 0.027342507615685463, | |
| "learning_rate": 7.560975609756099e-05, | |
| "loss": 0.0061, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 12.682926829268293, | |
| "grad_norm": 0.03514489158987999, | |
| "learning_rate": 7.317073170731707e-05, | |
| "loss": 0.006, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.926829268292684, | |
| "grad_norm": 0.02839997597038746, | |
| "learning_rate": 7.073170731707317e-05, | |
| "loss": 0.0056, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.170731707317072, | |
| "grad_norm": 0.025196045637130737, | |
| "learning_rate": 6.829268292682928e-05, | |
| "loss": 0.0057, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 13.414634146341463, | |
| "grad_norm": 0.022951927036046982, | |
| "learning_rate": 6.585365853658538e-05, | |
| "loss": 0.0053, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 13.658536585365853, | |
| "grad_norm": 0.02220899611711502, | |
| "learning_rate": 6.341463414634146e-05, | |
| "loss": 0.0056, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 13.902439024390244, | |
| "grad_norm": 0.026446521282196045, | |
| "learning_rate": 6.097560975609756e-05, | |
| "loss": 0.0051, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.146341463414634, | |
| "grad_norm": 0.021074431017041206, | |
| "learning_rate": 5.853658536585366e-05, | |
| "loss": 0.0051, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 14.390243902439025, | |
| "grad_norm": 0.026110034435987473, | |
| "learning_rate": 5.6097560975609764e-05, | |
| "loss": 0.0051, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 14.634146341463415, | |
| "grad_norm": 0.024351950734853745, | |
| "learning_rate": 5.365853658536586e-05, | |
| "loss": 0.0049, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.634146341463415, | |
| "eval_accuracy": 0.9642857142857143, | |
| "eval_loss": 0.15386557579040527, | |
| "eval_runtime": 24.339, | |
| "eval_samples_per_second": 5.752, | |
| "eval_steps_per_second": 0.74, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 14.878048780487806, | |
| "grad_norm": 0.020671434700489044, | |
| "learning_rate": 5.121951219512195e-05, | |
| "loss": 0.0048, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 15.121951219512194, | |
| "grad_norm": 0.02226124331355095, | |
| "learning_rate": 4.878048780487805e-05, | |
| "loss": 0.0046, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 15.365853658536585, | |
| "grad_norm": 0.02063142880797386, | |
| "learning_rate": 4.634146341463415e-05, | |
| "loss": 0.0046, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 15.609756097560975, | |
| "grad_norm": 0.02496950887143612, | |
| "learning_rate": 4.390243902439025e-05, | |
| "loss": 0.0048, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 15.853658536585366, | |
| "grad_norm": 0.023448029533028603, | |
| "learning_rate": 4.146341463414634e-05, | |
| "loss": 0.0046, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.097560975609756, | |
| "grad_norm": 0.02106618881225586, | |
| "learning_rate": 3.9024390243902444e-05, | |
| "loss": 0.0045, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 16.341463414634145, | |
| "grad_norm": 0.02371513471007347, | |
| "learning_rate": 3.6585365853658535e-05, | |
| "loss": 0.0043, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 16.585365853658537, | |
| "grad_norm": 0.02032136172056198, | |
| "learning_rate": 3.414634146341464e-05, | |
| "loss": 0.0044, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 16.829268292682926, | |
| "grad_norm": 0.020455192774534225, | |
| "learning_rate": 3.170731707317073e-05, | |
| "loss": 0.0043, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 17.073170731707318, | |
| "grad_norm": 0.023487282916903496, | |
| "learning_rate": 2.926829268292683e-05, | |
| "loss": 0.0044, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.073170731707318, | |
| "eval_accuracy": 0.9642857142857143, | |
| "eval_loss": 0.15085048973560333, | |
| "eval_runtime": 11.3536, | |
| "eval_samples_per_second": 12.331, | |
| "eval_steps_per_second": 1.585, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 17.317073170731707, | |
| "grad_norm": 0.018334366381168365, | |
| "learning_rate": 2.682926829268293e-05, | |
| "loss": 0.0044, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 17.5609756097561, | |
| "grad_norm": 0.01927388273179531, | |
| "learning_rate": 2.4390243902439026e-05, | |
| "loss": 0.0042, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 17.804878048780488, | |
| "grad_norm": 0.021918119862675667, | |
| "learning_rate": 2.1951219512195124e-05, | |
| "loss": 0.0042, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 18.048780487804876, | |
| "grad_norm": 0.019147571176290512, | |
| "learning_rate": 1.9512195121951222e-05, | |
| "loss": 0.0042, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 18.29268292682927, | |
| "grad_norm": 0.020710714161396027, | |
| "learning_rate": 1.707317073170732e-05, | |
| "loss": 0.0041, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 18.536585365853657, | |
| "grad_norm": 0.01830894872546196, | |
| "learning_rate": 1.4634146341463415e-05, | |
| "loss": 0.004, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 18.78048780487805, | |
| "grad_norm": 0.019078999757766724, | |
| "learning_rate": 1.2195121951219513e-05, | |
| "loss": 0.0042, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 19.024390243902438, | |
| "grad_norm": 0.020716849714517593, | |
| "learning_rate": 9.756097560975611e-06, | |
| "loss": 0.0042, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 19.26829268292683, | |
| "grad_norm": 0.018130987882614136, | |
| "learning_rate": 7.317073170731707e-06, | |
| "loss": 0.0041, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 19.51219512195122, | |
| "grad_norm": 0.018373191356658936, | |
| "learning_rate": 4.8780487804878055e-06, | |
| "loss": 0.0041, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.51219512195122, | |
| "eval_accuracy": 0.9642857142857143, | |
| "eval_loss": 0.14993587136268616, | |
| "eval_runtime": 14.7019, | |
| "eval_samples_per_second": 9.523, | |
| "eval_steps_per_second": 1.224, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 19.75609756097561, | |
| "grad_norm": 0.01835489086806774, | |
| "learning_rate": 2.4390243902439027e-06, | |
| "loss": 0.0039, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 0.02105889283120632, | |
| "learning_rate": 0.0, | |
| "loss": 0.0041, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 820, | |
| "total_flos": 1.004314187783209e+18, | |
| "train_loss": 0.1252097422029914, | |
| "train_runtime": 1111.906, | |
| "train_samples_per_second": 11.656, | |
| "train_steps_per_second": 0.737 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 820, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.004314187783209e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |