| { | |
| "best_metric": 0.8097881078720093, | |
| "best_model_checkpoint": "/media/mldrive/kcardenas/limb_classification_person_crop/beit-large-patch16-384/8_1e-5_5e-4_0.05/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": 1344286.0, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 1.5975, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2336448598130841, | |
| "grad_norm": 1232636.5, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.5637, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.35046728971962615, | |
| "grad_norm": 1367527.875, | |
| "learning_rate": 1.5e-06, | |
| "loss": 1.5906, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4672897196261682, | |
| "grad_norm": 1119946.625, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.5244, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5841121495327103, | |
| "grad_norm": 957303.1875, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.4616, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7009345794392523, | |
| "grad_norm": 554852.125, | |
| "learning_rate": 3e-06, | |
| "loss": 1.4292, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8177570093457944, | |
| "grad_norm": 931890.375, | |
| "learning_rate": 3.5e-06, | |
| "loss": 1.3964, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "grad_norm": 709493.4375, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.4037, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.2728026533996683, | |
| "eval_loss": 1.3978267908096313, | |
| "eval_runtime": 14.8259, | |
| "eval_samples_per_second": 81.344, | |
| "eval_steps_per_second": 2.563, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0514018691588785, | |
| "grad_norm": 470790.5, | |
| "learning_rate": 4.5e-06, | |
| "loss": 1.3872, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.1682242990654206, | |
| "grad_norm": 423176.3125, | |
| "learning_rate": 5e-06, | |
| "loss": 1.3614, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2850467289719627, | |
| "grad_norm": 504318.0, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 1.3493, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.4018691588785046, | |
| "grad_norm": 467144.0625, | |
| "learning_rate": 6e-06, | |
| "loss": 1.3507, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5186915887850467, | |
| "grad_norm": 794406.875, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 1.3447, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.6355140186915889, | |
| "grad_norm": 600493.1875, | |
| "learning_rate": 7e-06, | |
| "loss": 1.3342, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7523364485981308, | |
| "grad_norm": 503615.8125, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 1.3125, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 897004.6875, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2663, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.985981308411215, | |
| "grad_norm": 731026.0625, | |
| "learning_rate": 8.5e-06, | |
| "loss": 1.2477, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.494195688225539, | |
| "eval_loss": 1.1721693277359009, | |
| "eval_runtime": 14.8464, | |
| "eval_samples_per_second": 81.232, | |
| "eval_steps_per_second": 2.56, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.102803738317757, | |
| "grad_norm": 834097.25, | |
| "learning_rate": 9e-06, | |
| "loss": 1.2213, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2196261682242993, | |
| "grad_norm": 589023.0625, | |
| "learning_rate": 9.5e-06, | |
| "loss": 1.1893, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.336448598130841, | |
| "grad_norm": 431312.71875, | |
| "learning_rate": 1e-05, | |
| "loss": 1.2029, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.453271028037383, | |
| "grad_norm": 615542.5625, | |
| "learning_rate": 9.847560975609756e-06, | |
| "loss": 1.1831, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.5700934579439254, | |
| "grad_norm": 450194.375, | |
| "learning_rate": 9.695121951219513e-06, | |
| "loss": 1.1495, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6869158878504673, | |
| "grad_norm": 484475.25, | |
| "learning_rate": 9.542682926829268e-06, | |
| "loss": 1.1538, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.803738317757009, | |
| "grad_norm": 506196.09375, | |
| "learning_rate": 9.390243902439025e-06, | |
| "loss": 1.16, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9205607476635516, | |
| "grad_norm": 615921.1875, | |
| "learning_rate": 9.237804878048782e-06, | |
| "loss": 1.1297, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5970149253731343, | |
| "eval_loss": 0.9763943552970886, | |
| "eval_runtime": 14.869, | |
| "eval_samples_per_second": 81.109, | |
| "eval_steps_per_second": 2.556, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0373831775700935, | |
| "grad_norm": 449741.21875, | |
| "learning_rate": 9.085365853658538e-06, | |
| "loss": 1.1258, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.1542056074766354, | |
| "grad_norm": 588529.75, | |
| "learning_rate": 8.932926829268293e-06, | |
| "loss": 1.0668, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.2710280373831777, | |
| "grad_norm": 489126.1875, | |
| "learning_rate": 8.78048780487805e-06, | |
| "loss": 1.0899, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.3878504672897196, | |
| "grad_norm": 679644.9375, | |
| "learning_rate": 8.628048780487805e-06, | |
| "loss": 1.106, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.5046728971962615, | |
| "grad_norm": 469723.8125, | |
| "learning_rate": 8.475609756097562e-06, | |
| "loss": 1.0807, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.621495327102804, | |
| "grad_norm": 566260.4375, | |
| "learning_rate": 8.323170731707317e-06, | |
| "loss": 1.0741, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.7383177570093458, | |
| "grad_norm": 636508.4375, | |
| "learning_rate": 8.170731707317073e-06, | |
| "loss": 1.0789, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.8551401869158877, | |
| "grad_norm": 584515.9375, | |
| "learning_rate": 8.01829268292683e-06, | |
| "loss": 1.058, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.97196261682243, | |
| "grad_norm": 511375.34375, | |
| "learning_rate": 7.865853658536587e-06, | |
| "loss": 1.0621, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5912106135986733, | |
| "eval_loss": 1.0269614458084106, | |
| "eval_runtime": 15.3968, | |
| "eval_samples_per_second": 78.328, | |
| "eval_steps_per_second": 2.468, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.088785046728972, | |
| "grad_norm": 538369.125, | |
| "learning_rate": 7.713414634146342e-06, | |
| "loss": 1.0596, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.205607476635514, | |
| "grad_norm": 557098.75, | |
| "learning_rate": 7.560975609756098e-06, | |
| "loss": 1.0648, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.322429906542056, | |
| "grad_norm": 674418.9375, | |
| "learning_rate": 7.408536585365854e-06, | |
| "loss": 1.0151, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.4392523364485985, | |
| "grad_norm": 838321.1875, | |
| "learning_rate": 7.25609756097561e-06, | |
| "loss": 1.0045, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.55607476635514, | |
| "grad_norm": 595761.5625, | |
| "learning_rate": 7.1036585365853665e-06, | |
| "loss": 1.0178, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.672897196261682, | |
| "grad_norm": 595173.625, | |
| "learning_rate": 6.951219512195122e-06, | |
| "loss": 1.0028, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.789719626168225, | |
| "grad_norm": 766422.8125, | |
| "learning_rate": 6.798780487804879e-06, | |
| "loss": 1.0042, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.906542056074766, | |
| "grad_norm": 432231.9375, | |
| "learning_rate": 6.646341463414635e-06, | |
| "loss": 1.0244, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6625207296849088, | |
| "eval_loss": 0.9042626619338989, | |
| "eval_runtime": 14.8081, | |
| "eval_samples_per_second": 81.442, | |
| "eval_steps_per_second": 2.566, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0233644859813085, | |
| "grad_norm": 825474.0625, | |
| "learning_rate": 6.493902439024391e-06, | |
| "loss": 1.0353, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.140186915887851, | |
| "grad_norm": 628319.1875, | |
| "learning_rate": 6.341463414634147e-06, | |
| "loss": 0.9928, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.257009345794392, | |
| "grad_norm": 587063.25, | |
| "learning_rate": 6.189024390243903e-06, | |
| "loss": 0.9825, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 5.373831775700935, | |
| "grad_norm": 642682.375, | |
| "learning_rate": 6.0365853658536585e-06, | |
| "loss": 1.0014, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.490654205607477, | |
| "grad_norm": 723803.125, | |
| "learning_rate": 5.884146341463415e-06, | |
| "loss": 0.9637, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 5.607476635514018, | |
| "grad_norm": 466309.0625, | |
| "learning_rate": 5.731707317073171e-06, | |
| "loss": 0.9845, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.724299065420561, | |
| "grad_norm": 562217.5, | |
| "learning_rate": 5.579268292682928e-06, | |
| "loss": 1.0255, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 5.841121495327103, | |
| "grad_norm": 470339.15625, | |
| "learning_rate": 5.426829268292684e-06, | |
| "loss": 0.9714, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.957943925233645, | |
| "grad_norm": 607116.75, | |
| "learning_rate": 5.27439024390244e-06, | |
| "loss": 0.9623, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6824212271973465, | |
| "eval_loss": 0.8453999161720276, | |
| "eval_runtime": 14.9395, | |
| "eval_samples_per_second": 80.726, | |
| "eval_steps_per_second": 2.544, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.074766355140187, | |
| "grad_norm": 988413.125, | |
| "learning_rate": 5.121951219512195e-06, | |
| "loss": 0.9296, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.191588785046729, | |
| "grad_norm": 802783.125, | |
| "learning_rate": 4.9695121951219515e-06, | |
| "loss": 0.9449, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 6.308411214953271, | |
| "grad_norm": 591705.125, | |
| "learning_rate": 4.817073170731708e-06, | |
| "loss": 0.9578, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.425233644859813, | |
| "grad_norm": 625485.125, | |
| "learning_rate": 4.664634146341464e-06, | |
| "loss": 0.9165, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.542056074766355, | |
| "grad_norm": 682533.0, | |
| "learning_rate": 4.51219512195122e-06, | |
| "loss": 0.9226, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.658878504672897, | |
| "grad_norm": 809783.5, | |
| "learning_rate": 4.359756097560976e-06, | |
| "loss": 0.9039, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 6.775700934579439, | |
| "grad_norm": 544206.0, | |
| "learning_rate": 4.207317073170732e-06, | |
| "loss": 0.9374, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 6.892523364485982, | |
| "grad_norm": 631440.5625, | |
| "learning_rate": 4.054878048780488e-06, | |
| "loss": 0.8835, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.685737976782753, | |
| "eval_loss": 0.8481377363204956, | |
| "eval_runtime": 14.9433, | |
| "eval_samples_per_second": 80.705, | |
| "eval_steps_per_second": 2.543, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.009345794392523, | |
| "grad_norm": 695251.375, | |
| "learning_rate": 3.902439024390244e-06, | |
| "loss": 0.9291, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.126168224299065, | |
| "grad_norm": 729127.375, | |
| "learning_rate": 3.7500000000000005e-06, | |
| "loss": 0.8883, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 7.242990654205608, | |
| "grad_norm": 485646.09375, | |
| "learning_rate": 3.5975609756097562e-06, | |
| "loss": 0.9578, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.359813084112149, | |
| "grad_norm": 456878.625, | |
| "learning_rate": 3.4451219512195124e-06, | |
| "loss": 0.9008, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 7.4766355140186915, | |
| "grad_norm": 747060.8125, | |
| "learning_rate": 3.292682926829269e-06, | |
| "loss": 0.901, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.593457943925234, | |
| "grad_norm": 802504.8125, | |
| "learning_rate": 3.1402439024390246e-06, | |
| "loss": 0.9339, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 7.710280373831775, | |
| "grad_norm": 555262.9375, | |
| "learning_rate": 2.9878048780487808e-06, | |
| "loss": 0.9446, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 7.827102803738318, | |
| "grad_norm": 1096160.5, | |
| "learning_rate": 2.8353658536585365e-06, | |
| "loss": 0.8785, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 7.94392523364486, | |
| "grad_norm": 594814.5625, | |
| "learning_rate": 2.682926829268293e-06, | |
| "loss": 0.8853, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7114427860696517, | |
| "eval_loss": 0.8122716546058655, | |
| "eval_runtime": 15.1283, | |
| "eval_samples_per_second": 79.718, | |
| "eval_steps_per_second": 2.512, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.060747663551401, | |
| "grad_norm": 820075.875, | |
| "learning_rate": 2.530487804878049e-06, | |
| "loss": 0.8793, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 8.177570093457945, | |
| "grad_norm": 754132.3125, | |
| "learning_rate": 2.378048780487805e-06, | |
| "loss": 0.8494, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.294392523364486, | |
| "grad_norm": 783069.0, | |
| "learning_rate": 2.225609756097561e-06, | |
| "loss": 0.883, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 8.411214953271028, | |
| "grad_norm": 731905.0625, | |
| "learning_rate": 2.073170731707317e-06, | |
| "loss": 0.853, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.52803738317757, | |
| "grad_norm": 585234.9375, | |
| "learning_rate": 1.9207317073170733e-06, | |
| "loss": 0.8859, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 8.644859813084112, | |
| "grad_norm": 584208.0625, | |
| "learning_rate": 1.7682926829268294e-06, | |
| "loss": 0.8694, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.761682242990654, | |
| "grad_norm": 686404.625, | |
| "learning_rate": 1.6158536585365855e-06, | |
| "loss": 0.8791, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 8.878504672897197, | |
| "grad_norm": 625846.0625, | |
| "learning_rate": 1.4634146341463414e-06, | |
| "loss": 0.8576, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.995327102803738, | |
| "grad_norm": 641595.8125, | |
| "learning_rate": 1.3109756097560978e-06, | |
| "loss": 0.868, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.714759535655058, | |
| "eval_loss": 0.8097881078720093, | |
| "eval_runtime": 14.8393, | |
| "eval_samples_per_second": 81.271, | |
| "eval_steps_per_second": 2.561, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.11214953271028, | |
| "grad_norm": 709608.5625, | |
| "learning_rate": 1.158536585365854e-06, | |
| "loss": 0.8316, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.228971962616823, | |
| "grad_norm": 639677.75, | |
| "learning_rate": 1.0060975609756098e-06, | |
| "loss": 0.9073, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 9.345794392523365, | |
| "grad_norm": 656041.0625, | |
| "learning_rate": 8.53658536585366e-07, | |
| "loss": 0.8598, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.462616822429906, | |
| "grad_norm": 579136.0, | |
| "learning_rate": 7.012195121951221e-07, | |
| "loss": 0.7945, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 9.57943925233645, | |
| "grad_norm": 787287.3125, | |
| "learning_rate": 5.487804878048781e-07, | |
| "loss": 0.9198, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.69626168224299, | |
| "grad_norm": 813869.8125, | |
| "learning_rate": 3.963414634146342e-07, | |
| "loss": 0.8805, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 9.813084112149532, | |
| "grad_norm": 734961.1875, | |
| "learning_rate": 2.439024390243903e-07, | |
| "loss": 0.8613, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.929906542056075, | |
| "grad_norm": 704054.75, | |
| "learning_rate": 9.146341463414634e-08, | |
| "loss": 0.8474, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7072968490878938, | |
| "eval_loss": 0.8144016265869141, | |
| "eval_runtime": 15.0203, | |
| "eval_samples_per_second": 80.292, | |
| "eval_steps_per_second": 2.53, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2140, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_loss": 1.0635530485171023, | |
| "train_runtime": 1931.0211, | |
| "train_samples_per_second": 35.37, | |
| "train_steps_per_second": 1.108 | |
| } | |
| ], | |
| "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 | |
| } | |