| { | |
| "best_metric": 1.040339469909668, | |
| "best_model_checkpoint": "/media/mldrive/kcardenas/limb_classification_person_crop/beit-large-patch16-384/8_5e-5_5e-4_0.15/checkpoint-428", | |
| "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": 1226482.25, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.5898, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2336448598130841, | |
| "grad_norm": 883115.375, | |
| "learning_rate": 5e-06, | |
| "loss": 1.5063, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35046728971962615, | |
| "grad_norm": 729111.75, | |
| "learning_rate": 7.5e-06, | |
| "loss": 1.4618, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4672897196261682, | |
| "grad_norm": 401205.5, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4029, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5841121495327103, | |
| "grad_norm": 392407.21875, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.3853, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7009345794392523, | |
| "grad_norm": 296861.40625, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.3715, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8177570093457944, | |
| "grad_norm": 644126.5, | |
| "learning_rate": 1.75e-05, | |
| "loss": 1.344, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "grad_norm": 561771.25, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3404, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.39800995024875624, | |
| "eval_loss": 1.4185985326766968, | |
| "eval_runtime": 15.0252, | |
| "eval_samples_per_second": 80.265, | |
| "eval_steps_per_second": 2.529, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0514018691588785, | |
| "grad_norm": 450167.59375, | |
| "learning_rate": 2.25e-05, | |
| "loss": 1.2961, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.1682242990654206, | |
| "grad_norm": 275605.5625, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.2836, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2850467289719627, | |
| "grad_norm": 417004.96875, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 1.2374, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.4018691588785046, | |
| "grad_norm": 313505.09375, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2688, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5186915887850467, | |
| "grad_norm": 345996.875, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 1.2501, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.6355140186915889, | |
| "grad_norm": 439750.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.2455, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7523364485981308, | |
| "grad_norm": 192930.5625, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.2433, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 403192.8125, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1843, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.985981308411215, | |
| "grad_norm": 312608.1875, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.1899, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6442786069651741, | |
| "eval_loss": 1.040339469909668, | |
| "eval_runtime": 15.0124, | |
| "eval_samples_per_second": 80.333, | |
| "eval_steps_per_second": 2.531, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.102803738317757, | |
| "grad_norm": 565702.9375, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1768, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2196261682242993, | |
| "grad_norm": 228019.625, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.3692, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.336448598130841, | |
| "grad_norm": 97234.1796875, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4012, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.453271028037383, | |
| "grad_norm": 198352.921875, | |
| "learning_rate": 4.923780487804878e-05, | |
| "loss": 1.392, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.5700934579439254, | |
| "grad_norm": 124283.6953125, | |
| "learning_rate": 4.847560975609756e-05, | |
| "loss": 1.3919, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6869158878504673, | |
| "grad_norm": 120723.7578125, | |
| "learning_rate": 4.771341463414634e-05, | |
| "loss": 1.3871, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.803738317757009, | |
| "grad_norm": 213803.59375, | |
| "learning_rate": 4.695121951219512e-05, | |
| "loss": 1.3944, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9205607476635516, | |
| "grad_norm": 286424.5625, | |
| "learning_rate": 4.618902439024391e-05, | |
| "loss": 1.3962, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.2744610281923715, | |
| "eval_loss": 1.3858356475830078, | |
| "eval_runtime": 15.0104, | |
| "eval_samples_per_second": 80.344, | |
| "eval_steps_per_second": 2.532, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0373831775700935, | |
| "grad_norm": 211193.328125, | |
| "learning_rate": 4.542682926829269e-05, | |
| "loss": 1.3901, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.1542056074766354, | |
| "grad_norm": 111855.8359375, | |
| "learning_rate": 4.466463414634147e-05, | |
| "loss": 1.3935, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.2710280373831777, | |
| "grad_norm": 157092.0625, | |
| "learning_rate": 4.390243902439025e-05, | |
| "loss": 1.3811, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.3878504672897196, | |
| "grad_norm": 127590.5703125, | |
| "learning_rate": 4.314024390243903e-05, | |
| "loss": 1.3921, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.5046728971962615, | |
| "grad_norm": 100796.25, | |
| "learning_rate": 4.237804878048781e-05, | |
| "loss": 1.3852, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.621495327102804, | |
| "grad_norm": 139129.4375, | |
| "learning_rate": 4.161585365853659e-05, | |
| "loss": 1.3856, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.7383177570093458, | |
| "grad_norm": 176988.171875, | |
| "learning_rate": 4.085365853658537e-05, | |
| "loss": 1.3844, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.8551401869158877, | |
| "grad_norm": 127246.34375, | |
| "learning_rate": 4.0091463414634153e-05, | |
| "loss": 1.3805, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.97196261682243, | |
| "grad_norm": 212300.828125, | |
| "learning_rate": 3.932926829268293e-05, | |
| "loss": 1.3914, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.2744610281923715, | |
| "eval_loss": 1.3880783319473267, | |
| "eval_runtime": 15.2034, | |
| "eval_samples_per_second": 79.324, | |
| "eval_steps_per_second": 2.499, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.088785046728972, | |
| "grad_norm": 61669.88671875, | |
| "learning_rate": 3.856707317073171e-05, | |
| "loss": 1.3905, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.205607476635514, | |
| "grad_norm": 223599.234375, | |
| "learning_rate": 3.780487804878049e-05, | |
| "loss": 1.3906, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.322429906542056, | |
| "grad_norm": 247944.15625, | |
| "learning_rate": 3.704268292682927e-05, | |
| "loss": 1.3902, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.4392523364485985, | |
| "grad_norm": 136334.375, | |
| "learning_rate": 3.628048780487805e-05, | |
| "loss": 1.3901, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.55607476635514, | |
| "grad_norm": 161834.765625, | |
| "learning_rate": 3.551829268292683e-05, | |
| "loss": 1.3842, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.672897196261682, | |
| "grad_norm": 160751.734375, | |
| "learning_rate": 3.475609756097561e-05, | |
| "loss": 1.3856, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.789719626168225, | |
| "grad_norm": 152146.84375, | |
| "learning_rate": 3.399390243902439e-05, | |
| "loss": 1.3823, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.906542056074766, | |
| "grad_norm": 193127.203125, | |
| "learning_rate": 3.323170731707317e-05, | |
| "loss": 1.3868, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.27611940298507465, | |
| "eval_loss": 1.3848323822021484, | |
| "eval_runtime": 15.4933, | |
| "eval_samples_per_second": 77.84, | |
| "eval_steps_per_second": 2.453, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0233644859813085, | |
| "grad_norm": 238072.984375, | |
| "learning_rate": 3.246951219512195e-05, | |
| "loss": 1.3888, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.140186915887851, | |
| "grad_norm": 133656.28125, | |
| "learning_rate": 3.170731707317073e-05, | |
| "loss": 1.3863, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.257009345794392, | |
| "grad_norm": 120263.9453125, | |
| "learning_rate": 3.094512195121951e-05, | |
| "loss": 1.3856, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 5.373831775700935, | |
| "grad_norm": 167475.484375, | |
| "learning_rate": 3.0182926829268294e-05, | |
| "loss": 1.3885, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.490654205607477, | |
| "grad_norm": 197263.59375, | |
| "learning_rate": 2.9420731707317074e-05, | |
| "loss": 1.385, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 5.607476635514018, | |
| "grad_norm": 145337.453125, | |
| "learning_rate": 2.8658536585365854e-05, | |
| "loss": 1.3807, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.724299065420561, | |
| "grad_norm": 119471.03125, | |
| "learning_rate": 2.7896341463414637e-05, | |
| "loss": 1.3806, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 5.841121495327103, | |
| "grad_norm": 139454.765625, | |
| "learning_rate": 2.7134146341463417e-05, | |
| "loss": 1.3843, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.957943925233645, | |
| "grad_norm": 147728.828125, | |
| "learning_rate": 2.6371951219512197e-05, | |
| "loss": 1.3895, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.2744610281923715, | |
| "eval_loss": 1.386699914932251, | |
| "eval_runtime": 14.8197, | |
| "eval_samples_per_second": 81.378, | |
| "eval_steps_per_second": 2.564, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.074766355140187, | |
| "grad_norm": 217932.9375, | |
| "learning_rate": 2.5609756097560977e-05, | |
| "loss": 1.3858, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.191588785046729, | |
| "grad_norm": 103804.140625, | |
| "learning_rate": 2.4847560975609756e-05, | |
| "loss": 1.3839, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 6.308411214953271, | |
| "grad_norm": 144548.296875, | |
| "learning_rate": 2.4085365853658536e-05, | |
| "loss": 1.3757, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.425233644859813, | |
| "grad_norm": 116841.2421875, | |
| "learning_rate": 2.332317073170732e-05, | |
| "loss": 1.3832, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.542056074766355, | |
| "grad_norm": 60290.0859375, | |
| "learning_rate": 2.25609756097561e-05, | |
| "loss": 1.3866, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.658878504672897, | |
| "grad_norm": 142147.796875, | |
| "learning_rate": 2.179878048780488e-05, | |
| "loss": 1.3812, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 6.775700934579439, | |
| "grad_norm": 77919.5390625, | |
| "learning_rate": 2.103658536585366e-05, | |
| "loss": 1.3852, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 6.892523364485982, | |
| "grad_norm": 112284.5234375, | |
| "learning_rate": 2.0274390243902442e-05, | |
| "loss": 1.3871, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.2744610281923715, | |
| "eval_loss": 1.3830299377441406, | |
| "eval_runtime": 15.5048, | |
| "eval_samples_per_second": 77.782, | |
| "eval_steps_per_second": 2.451, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.009345794392523, | |
| "grad_norm": 219239.40625, | |
| "learning_rate": 1.9512195121951222e-05, | |
| "loss": 1.3871, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.126168224299065, | |
| "grad_norm": 139572.75, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 1.3849, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 7.242990654205608, | |
| "grad_norm": 71637.71875, | |
| "learning_rate": 1.798780487804878e-05, | |
| "loss": 1.3832, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.359813084112149, | |
| "grad_norm": 134367.84375, | |
| "learning_rate": 1.722560975609756e-05, | |
| "loss": 1.3789, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 7.4766355140186915, | |
| "grad_norm": 146302.65625, | |
| "learning_rate": 1.6463414634146345e-05, | |
| "loss": 1.3838, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.593457943925234, | |
| "grad_norm": 65325.31640625, | |
| "learning_rate": 1.5701219512195124e-05, | |
| "loss": 1.3808, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 7.710280373831775, | |
| "grad_norm": 105734.6328125, | |
| "learning_rate": 1.4939024390243902e-05, | |
| "loss": 1.3811, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 7.827102803738318, | |
| "grad_norm": 158432.8125, | |
| "learning_rate": 1.4176829268292682e-05, | |
| "loss": 1.3782, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 7.94392523364486, | |
| "grad_norm": 109856.3671875, | |
| "learning_rate": 1.3414634146341466e-05, | |
| "loss": 1.3802, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.29270315091210614, | |
| "eval_loss": 1.378788709640503, | |
| "eval_runtime": 15.2702, | |
| "eval_samples_per_second": 78.977, | |
| "eval_steps_per_second": 2.489, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.060747663551401, | |
| "grad_norm": 131191.734375, | |
| "learning_rate": 1.2652439024390245e-05, | |
| "loss": 1.3796, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 8.177570093457945, | |
| "grad_norm": 111280.453125, | |
| "learning_rate": 1.1890243902439025e-05, | |
| "loss": 1.3764, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.294392523364486, | |
| "grad_norm": 53637.3828125, | |
| "learning_rate": 1.1128048780487805e-05, | |
| "loss": 1.3815, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 8.411214953271028, | |
| "grad_norm": 67744.1640625, | |
| "learning_rate": 1.0365853658536585e-05, | |
| "loss": 1.3667, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.52803738317757, | |
| "grad_norm": 88018.4921875, | |
| "learning_rate": 9.603658536585366e-06, | |
| "loss": 1.369, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 8.644859813084112, | |
| "grad_norm": 92662.328125, | |
| "learning_rate": 8.841463414634146e-06, | |
| "loss": 1.3732, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.761682242990654, | |
| "grad_norm": 146536.6875, | |
| "learning_rate": 8.079268292682928e-06, | |
| "loss": 1.3732, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 8.878504672897197, | |
| "grad_norm": 132991.578125, | |
| "learning_rate": 7.317073170731707e-06, | |
| "loss": 1.3685, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.995327102803738, | |
| "grad_norm": 134634.984375, | |
| "learning_rate": 6.554878048780488e-06, | |
| "loss": 1.3745, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.3200663349917081, | |
| "eval_loss": 1.367501974105835, | |
| "eval_runtime": 15.3432, | |
| "eval_samples_per_second": 78.602, | |
| "eval_steps_per_second": 2.477, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.11214953271028, | |
| "grad_norm": 111594.0234375, | |
| "learning_rate": 5.792682926829269e-06, | |
| "loss": 1.3629, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.228971962616823, | |
| "grad_norm": 260942.34375, | |
| "learning_rate": 5.030487804878049e-06, | |
| "loss": 1.3681, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 9.345794392523365, | |
| "grad_norm": 132676.8125, | |
| "learning_rate": 4.26829268292683e-06, | |
| "loss": 1.3568, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.462616822429906, | |
| "grad_norm": 137239.046875, | |
| "learning_rate": 3.5060975609756102e-06, | |
| "loss": 1.3566, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 9.57943925233645, | |
| "grad_norm": 167074.9375, | |
| "learning_rate": 2.7439024390243905e-06, | |
| "loss": 1.3556, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.69626168224299, | |
| "grad_norm": 89201.1015625, | |
| "learning_rate": 1.9817073170731707e-06, | |
| "loss": 1.3697, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 9.813084112149532, | |
| "grad_norm": 104514.7109375, | |
| "learning_rate": 1.2195121951219514e-06, | |
| "loss": 1.346, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.929906542056075, | |
| "grad_norm": 137668.4375, | |
| "learning_rate": 4.573170731707317e-07, | |
| "loss": 1.356, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.3383084577114428, | |
| "eval_loss": 1.3566746711730957, | |
| "eval_runtime": 14.6136, | |
| "eval_samples_per_second": 82.526, | |
| "eval_steps_per_second": 2.6, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2140, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_loss": 1.3681953902556518, | |
| "train_runtime": 1910.7588, | |
| "train_samples_per_second": 35.745, | |
| "train_steps_per_second": 1.12 | |
| } | |
| ], | |
| "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 | |
| } | |