| { | |
| "best_metric": 0.8748477697372437, | |
| "best_model_checkpoint": "/media/mldrive/kcardenas/limb_classification_person_crop/beit-large-patch16-384/8_1e-5_1e-4_0.1/checkpoint-1926", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 2140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11682242990654206, | |
| "grad_norm": 1288282.125, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.6003, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2336448598130841, | |
| "grad_norm": 1179080.75, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.5656, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35046728971962615, | |
| "grad_norm": 1302664.0, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.5881, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4672897196261682, | |
| "grad_norm": 1064078.25, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.5212, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5841121495327103, | |
| "grad_norm": 891409.25, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.4607, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7009345794392523, | |
| "grad_norm": 506389.46875, | |
| "learning_rate": 3e-06, | |
| "loss": 1.4285, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8177570093457944, | |
| "grad_norm": 858052.0, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.3972, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "grad_norm": 647255.5, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.4023, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.2744610281923715, | |
| "eval_loss": 1.3974584341049194, | |
| "eval_runtime": 15.3016, | |
| "eval_samples_per_second": 78.815, | |
| "eval_steps_per_second": 2.483, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0514018691588785, | |
| "grad_norm": 432902.65625, | |
| "learning_rate": 4.5e-06, | |
| "loss": 1.3877, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.1682242990654206, | |
| "grad_norm": 387082.78125, | |
| "learning_rate": 5e-06, | |
| "loss": 1.3642, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2850467289719627, | |
| "grad_norm": 460995.21875, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 1.3537, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.4018691588785046, | |
| "grad_norm": 426754.125, | |
| "learning_rate": 6e-06, | |
| "loss": 1.3536, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5186915887850467, | |
| "grad_norm": 689839.1875, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.3486, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.6355140186915889, | |
| "grad_norm": 525583.1875, | |
| "learning_rate": 7e-06, | |
| "loss": 1.3369, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7523364485981308, | |
| "grad_norm": 458157.90625, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.3238, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 716735.5, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2738, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.985981308411215, | |
| "grad_norm": 547995.0625, | |
| "learning_rate": 8.5e-06, | |
| "loss": 1.2554, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.49087893864013266, | |
| "eval_loss": 1.1979795694351196, | |
| "eval_runtime": 15.2164, | |
| "eval_samples_per_second": 79.257, | |
| "eval_steps_per_second": 2.497, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.102803738317757, | |
| "grad_norm": 715070.1875, | |
| "learning_rate": 9e-06, | |
| "loss": 1.2346, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2196261682242993, | |
| "grad_norm": 568589.125, | |
| "learning_rate": 9.5e-06, | |
| "loss": 1.2079, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.336448598130841, | |
| "grad_norm": 374537.625, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2209, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.453271028037383, | |
| "grad_norm": 539741.6875, | |
| "learning_rate": 9.847560975609756e-06, | |
| "loss": 1.2072, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.5700934579439254, | |
| "grad_norm": 434154.53125, | |
| "learning_rate": 9.695121951219513e-06, | |
| "loss": 1.1796, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6869158878504673, | |
| "grad_norm": 416858.96875, | |
| "learning_rate": 9.542682926829268e-06, | |
| "loss": 1.182, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.803738317757009, | |
| "grad_norm": 434543.6875, | |
| "learning_rate": 9.390243902439025e-06, | |
| "loss": 1.1822, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9205607476635516, | |
| "grad_norm": 560064.375, | |
| "learning_rate": 9.237804878048782e-06, | |
| "loss": 1.1636, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5978441127694859, | |
| "eval_loss": 1.027214765548706, | |
| "eval_runtime": 15.3221, | |
| "eval_samples_per_second": 78.71, | |
| "eval_steps_per_second": 2.48, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0373831775700935, | |
| "grad_norm": 413180.0, | |
| "learning_rate": 9.085365853658538e-06, | |
| "loss": 1.1574, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.1542056074766354, | |
| "grad_norm": 562978.6875, | |
| "learning_rate": 8.932926829268293e-06, | |
| "loss": 1.0971, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.2710280373831777, | |
| "grad_norm": 429750.75, | |
| "learning_rate": 8.78048780487805e-06, | |
| "loss": 1.1231, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.3878504672897196, | |
| "grad_norm": 592197.9375, | |
| "learning_rate": 8.628048780487805e-06, | |
| "loss": 1.1382, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.5046728971962615, | |
| "grad_norm": 402836.8125, | |
| "learning_rate": 8.475609756097562e-06, | |
| "loss": 1.1188, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.621495327102804, | |
| "grad_norm": 506094.375, | |
| "learning_rate": 8.323170731707317e-06, | |
| "loss": 1.1118, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.7383177570093458, | |
| "grad_norm": 520306.875, | |
| "learning_rate": 8.170731707317073e-06, | |
| "loss": 1.1209, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.8551401869158877, | |
| "grad_norm": 517659.96875, | |
| "learning_rate": 8.01829268292683e-06, | |
| "loss": 1.0998, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.97196261682243, | |
| "grad_norm": 463775.65625, | |
| "learning_rate": 7.865853658536587e-06, | |
| "loss": 1.0988, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5953565505804311, | |
| "eval_loss": 1.0476573705673218, | |
| "eval_runtime": 15.1491, | |
| "eval_samples_per_second": 79.609, | |
| "eval_steps_per_second": 2.508, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.088785046728972, | |
| "grad_norm": 501856.03125, | |
| "learning_rate": 7.713414634146342e-06, | |
| "loss": 1.0932, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.205607476635514, | |
| "grad_norm": 517486.46875, | |
| "learning_rate": 7.560975609756098e-06, | |
| "loss": 1.103, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.322429906542056, | |
| "grad_norm": 595251.0, | |
| "learning_rate": 7.408536585365854e-06, | |
| "loss": 1.0561, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.4392523364485985, | |
| "grad_norm": 681244.0625, | |
| "learning_rate": 7.25609756097561e-06, | |
| "loss": 1.0518, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.55607476635514, | |
| "grad_norm": 482226.96875, | |
| "learning_rate": 7.1036585365853665e-06, | |
| "loss": 1.0594, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.672897196261682, | |
| "grad_norm": 491086.5625, | |
| "learning_rate": 6.951219512195122e-06, | |
| "loss": 1.0495, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.789719626168225, | |
| "grad_norm": 682658.25, | |
| "learning_rate": 6.798780487804879e-06, | |
| "loss": 1.05, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.906542056074766, | |
| "grad_norm": 371352.9375, | |
| "learning_rate": 6.646341463414635e-06, | |
| "loss": 1.068, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6592039800995025, | |
| "eval_loss": 0.9599226713180542, | |
| "eval_runtime": 15.1792, | |
| "eval_samples_per_second": 79.451, | |
| "eval_steps_per_second": 2.503, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0233644859813085, | |
| "grad_norm": 737796.0625, | |
| "learning_rate": 6.493902439024391e-06, | |
| "loss": 1.0747, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.140186915887851, | |
| "grad_norm": 544087.5625, | |
| "learning_rate": 6.341463414634147e-06, | |
| "loss": 1.042, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.257009345794392, | |
| "grad_norm": 505016.96875, | |
| "learning_rate": 6.189024390243903e-06, | |
| "loss": 1.0356, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 5.373831775700935, | |
| "grad_norm": 552820.9375, | |
| "learning_rate": 6.0365853658536585e-06, | |
| "loss": 1.0499, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.490654205607477, | |
| "grad_norm": 599422.0625, | |
| "learning_rate": 5.884146341463415e-06, | |
| "loss": 1.0164, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 5.607476635514018, | |
| "grad_norm": 393778.09375, | |
| "learning_rate": 5.731707317073171e-06, | |
| "loss": 1.0345, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.724299065420561, | |
| "grad_norm": 477965.125, | |
| "learning_rate": 5.579268292682928e-06, | |
| "loss": 1.0708, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 5.841121495327103, | |
| "grad_norm": 424191.53125, | |
| "learning_rate": 5.426829268292684e-06, | |
| "loss": 1.0142, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.957943925233645, | |
| "grad_norm": 540424.5625, | |
| "learning_rate": 5.27439024390244e-06, | |
| "loss": 1.0159, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6807628524046434, | |
| "eval_loss": 0.9090853333473206, | |
| "eval_runtime": 15.2965, | |
| "eval_samples_per_second": 78.842, | |
| "eval_steps_per_second": 2.484, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.074766355140187, | |
| "grad_norm": 677101.5, | |
| "learning_rate": 5.121951219512195e-06, | |
| "loss": 0.9873, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.191588785046729, | |
| "grad_norm": 694898.8125, | |
| "learning_rate": 4.9695121951219515e-06, | |
| "loss": 1.0045, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 6.308411214953271, | |
| "grad_norm": 514024.75, | |
| "learning_rate": 4.817073170731708e-06, | |
| "loss": 1.0045, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.425233644859813, | |
| "grad_norm": 590044.625, | |
| "learning_rate": 4.664634146341464e-06, | |
| "loss": 0.9731, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.542056074766355, | |
| "grad_norm": 611512.4375, | |
| "learning_rate": 4.51219512195122e-06, | |
| "loss": 0.9752, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.658878504672897, | |
| "grad_norm": 688492.6875, | |
| "learning_rate": 4.359756097560976e-06, | |
| "loss": 0.9653, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 6.775700934579439, | |
| "grad_norm": 507507.28125, | |
| "learning_rate": 4.207317073170732e-06, | |
| "loss": 0.9916, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 6.892523364485982, | |
| "grad_norm": 526057.875, | |
| "learning_rate": 4.054878048780488e-06, | |
| "loss": 0.9484, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6890547263681592, | |
| "eval_loss": 0.9250107407569885, | |
| "eval_runtime": 15.1589, | |
| "eval_samples_per_second": 79.557, | |
| "eval_steps_per_second": 2.507, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.009345794392523, | |
| "grad_norm": 584858.375, | |
| "learning_rate": 3.902439024390244e-06, | |
| "loss": 0.9819, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.126168224299065, | |
| "grad_norm": 642905.0, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.9513, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 7.242990654205608, | |
| "grad_norm": 444722.8125, | |
| "learning_rate": 3.5975609756097562e-06, | |
| "loss": 1.0087, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.359813084112149, | |
| "grad_norm": 418072.0, | |
| "learning_rate": 3.4451219512195124e-06, | |
| "loss": 0.9674, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 7.4766355140186915, | |
| "grad_norm": 616286.125, | |
| "learning_rate": 3.292682926829269e-06, | |
| "loss": 0.9598, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.593457943925234, | |
| "grad_norm": 660370.9375, | |
| "learning_rate": 3.1402439024390246e-06, | |
| "loss": 0.9899, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 7.710280373831775, | |
| "grad_norm": 474009.0, | |
| "learning_rate": 2.9878048780487808e-06, | |
| "loss": 0.9953, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 7.827102803738318, | |
| "grad_norm": 893681.125, | |
| "learning_rate": 2.8353658536585365e-06, | |
| "loss": 0.9445, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 7.94392523364486, | |
| "grad_norm": 500253.84375, | |
| "learning_rate": 2.682926829268293e-06, | |
| "loss": 0.9464, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.714759535655058, | |
| "eval_loss": 0.8801141977310181, | |
| "eval_runtime": 14.8458, | |
| "eval_samples_per_second": 81.235, | |
| "eval_steps_per_second": 2.56, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.060747663551401, | |
| "grad_norm": 714507.3125, | |
| "learning_rate": 2.530487804878049e-06, | |
| "loss": 0.941, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 8.177570093457945, | |
| "grad_norm": 655121.4375, | |
| "learning_rate": 2.378048780487805e-06, | |
| "loss": 0.9218, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.294392523364486, | |
| "grad_norm": 656624.75, | |
| "learning_rate": 2.225609756097561e-06, | |
| "loss": 0.9422, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 8.411214953271028, | |
| "grad_norm": 643071.5, | |
| "learning_rate": 2.073170731707317e-06, | |
| "loss": 0.9206, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.52803738317757, | |
| "grad_norm": 470872.96875, | |
| "learning_rate": 1.9207317073170733e-06, | |
| "loss": 0.9458, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 8.644859813084112, | |
| "grad_norm": 497745.5, | |
| "learning_rate": 1.7682926829268294e-06, | |
| "loss": 0.9301, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.761682242990654, | |
| "grad_norm": 596386.875, | |
| "learning_rate": 1.6158536585365855e-06, | |
| "loss": 0.9394, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 8.878504672897197, | |
| "grad_norm": 537784.75, | |
| "learning_rate": 1.4634146341463414e-06, | |
| "loss": 0.9246, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.995327102803738, | |
| "grad_norm": 554460.5625, | |
| "learning_rate": 1.3109756097560978e-06, | |
| "loss": 0.9361, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7155887230514096, | |
| "eval_loss": 0.8748477697372437, | |
| "eval_runtime": 15.4428, | |
| "eval_samples_per_second": 78.094, | |
| "eval_steps_per_second": 2.461, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.11214953271028, | |
| "grad_norm": 641775.6875, | |
| "learning_rate": 1.158536585365854e-06, | |
| "loss": 0.9021, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.228971962616823, | |
| "grad_norm": 552990.4375, | |
| "learning_rate": 1.0060975609756098e-06, | |
| "loss": 0.9592, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 9.345794392523365, | |
| "grad_norm": 560462.5, | |
| "learning_rate": 8.53658536585366e-07, | |
| "loss": 0.924, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.462616822429906, | |
| "grad_norm": 517514.09375, | |
| "learning_rate": 7.012195121951221e-07, | |
| "loss": 0.8702, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 9.57943925233645, | |
| "grad_norm": 668150.0, | |
| "learning_rate": 5.487804878048781e-07, | |
| "loss": 0.9793, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.69626168224299, | |
| "grad_norm": 738988.9375, | |
| "learning_rate": 3.963414634146342e-07, | |
| "loss": 0.9452, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 9.813084112149532, | |
| "grad_norm": 622479.125, | |
| "learning_rate": 2.439024390243903e-07, | |
| "loss": 0.9314, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.929906542056075, | |
| "grad_norm": 633154.1875, | |
| "learning_rate": 9.146341463414634e-08, | |
| "loss": 0.919, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7056384742951907, | |
| "eval_loss": 0.8807059526443481, | |
| "eval_runtime": 15.7901, | |
| "eval_samples_per_second": 76.377, | |
| "eval_steps_per_second": 2.407, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2140, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_loss": 1.103826135118431, | |
| "train_runtime": 1935.7344, | |
| "train_samples_per_second": 35.284, | |
| "train_steps_per_second": 1.106 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |