| { | |
| "best_metric": 0.8566215634346008, | |
| "best_model_checkpoint": "/media/mldrive/kcardenas/limb_classification_person_crop/beit-large-patch16-384/8_5e-5_5e-3_0.15/checkpoint-1712", | |
| "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": 729109.375, | |
| "learning_rate": 7.5e-06, | |
| "loss": 1.4618, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.4672897196261682, | |
| "grad_norm": 401195.65625, | |
| "learning_rate": 1e-05, | |
| "loss": 1.4029, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.5841121495327103, | |
| "grad_norm": 392395.125, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.3853, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.7009345794392523, | |
| "grad_norm": 296917.9375, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.3715, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.8177570093457944, | |
| "grad_norm": 643953.25, | |
| "learning_rate": 1.75e-05, | |
| "loss": 1.3439, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.9345794392523364, | |
| "grad_norm": 562499.125, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3405, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4013266998341625, | |
| "eval_loss": 1.4132930040359497, | |
| "eval_runtime": 15.7147, | |
| "eval_samples_per_second": 76.743, | |
| "eval_steps_per_second": 2.418, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 1.0514018691588785, | |
| "grad_norm": 438941.78125, | |
| "learning_rate": 2.25e-05, | |
| "loss": 1.296, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 1.1682242990654206, | |
| "grad_norm": 273291.78125, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.2843, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.2850467289719627, | |
| "grad_norm": 376603.28125, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 1.2351, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 1.4018691588785046, | |
| "grad_norm": 292181.65625, | |
| "learning_rate": 3e-05, | |
| "loss": 1.2693, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.5186915887850467, | |
| "grad_norm": 360521.46875, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 1.2471, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 1.6355140186915889, | |
| "grad_norm": 379625.5625, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.2438, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.7523364485981308, | |
| "grad_norm": 204172.15625, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.2378, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.8691588785046729, | |
| "grad_norm": 348753.78125, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1822, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.985981308411215, | |
| "grad_norm": 324683.03125, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.1968, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6243781094527363, | |
| "eval_loss": 1.0506110191345215, | |
| "eval_runtime": 14.8511, | |
| "eval_samples_per_second": 81.206, | |
| "eval_steps_per_second": 2.559, | |
| "step": 428 | |
| }, | |
| { | |
| "epoch": 2.102803738317757, | |
| "grad_norm": 455549.875, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1799, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.2196261682242993, | |
| "grad_norm": 439833.65625, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.1849, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 2.336448598130841, | |
| "grad_norm": 229161.1875, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1697, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.453271028037383, | |
| "grad_norm": 420862.84375, | |
| "learning_rate": 4.923780487804878e-05, | |
| "loss": 1.1712, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 2.5700934579439254, | |
| "grad_norm": 167969.25, | |
| "learning_rate": 4.847560975609756e-05, | |
| "loss": 1.161, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.6869158878504673, | |
| "grad_norm": 270386.03125, | |
| "learning_rate": 4.771341463414634e-05, | |
| "loss": 1.1955, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 2.803738317757009, | |
| "grad_norm": 209195.28125, | |
| "learning_rate": 4.695121951219512e-05, | |
| "loss": 1.1484, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 2.9205607476635516, | |
| "grad_norm": 259627.890625, | |
| "learning_rate": 4.618902439024391e-05, | |
| "loss": 1.1686, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6401326699834162, | |
| "eval_loss": 1.0164262056350708, | |
| "eval_runtime": 15.4208, | |
| "eval_samples_per_second": 78.206, | |
| "eval_steps_per_second": 2.464, | |
| "step": 642 | |
| }, | |
| { | |
| "epoch": 3.0373831775700935, | |
| "grad_norm": 315698.75, | |
| "learning_rate": 4.542682926829269e-05, | |
| "loss": 1.1514, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.1542056074766354, | |
| "grad_norm": 282164.09375, | |
| "learning_rate": 4.466463414634147e-05, | |
| "loss": 1.1035, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 3.2710280373831777, | |
| "grad_norm": 293740.1875, | |
| "learning_rate": 4.390243902439025e-05, | |
| "loss": 1.1259, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 3.3878504672897196, | |
| "grad_norm": 235999.71875, | |
| "learning_rate": 4.314024390243903e-05, | |
| "loss": 1.1469, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 3.5046728971962615, | |
| "grad_norm": 201727.953125, | |
| "learning_rate": 4.237804878048781e-05, | |
| "loss": 1.1153, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 3.621495327102804, | |
| "grad_norm": 293322.71875, | |
| "learning_rate": 4.161585365853659e-05, | |
| "loss": 1.1031, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 3.7383177570093458, | |
| "grad_norm": 275993.9375, | |
| "learning_rate": 4.085365853658537e-05, | |
| "loss": 1.1286, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 3.8551401869158877, | |
| "grad_norm": 240362.65625, | |
| "learning_rate": 4.0091463414634153e-05, | |
| "loss": 1.1203, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 3.97196261682243, | |
| "grad_norm": 321376.5, | |
| "learning_rate": 3.932926829268293e-05, | |
| "loss": 1.0942, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6608623548922057, | |
| "eval_loss": 0.9993247389793396, | |
| "eval_runtime": 15.7565, | |
| "eval_samples_per_second": 76.54, | |
| "eval_steps_per_second": 2.412, | |
| "step": 856 | |
| }, | |
| { | |
| "epoch": 4.088785046728972, | |
| "grad_norm": 304064.4375, | |
| "learning_rate": 3.856707317073171e-05, | |
| "loss": 1.0738, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 4.205607476635514, | |
| "grad_norm": 234398.5625, | |
| "learning_rate": 3.780487804878049e-05, | |
| "loss": 1.0889, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 4.322429906542056, | |
| "grad_norm": 430511.1875, | |
| "learning_rate": 3.704268292682927e-05, | |
| "loss": 1.065, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 4.4392523364485985, | |
| "grad_norm": 320453.375, | |
| "learning_rate": 3.628048780487805e-05, | |
| "loss": 1.0622, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 4.55607476635514, | |
| "grad_norm": 242444.015625, | |
| "learning_rate": 3.551829268292683e-05, | |
| "loss": 1.0645, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 4.672897196261682, | |
| "grad_norm": 177630.03125, | |
| "learning_rate": 3.475609756097561e-05, | |
| "loss": 1.0619, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 4.789719626168225, | |
| "grad_norm": 320625.0, | |
| "learning_rate": 3.399390243902439e-05, | |
| "loss": 1.0476, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 4.906542056074766, | |
| "grad_norm": 156043.859375, | |
| "learning_rate": 3.323170731707317e-05, | |
| "loss": 1.0901, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6890547263681592, | |
| "eval_loss": 0.9591612219810486, | |
| "eval_runtime": 15.708, | |
| "eval_samples_per_second": 76.776, | |
| "eval_steps_per_second": 2.419, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 5.0233644859813085, | |
| "grad_norm": 356902.34375, | |
| "learning_rate": 3.246951219512195e-05, | |
| "loss": 1.083, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 5.140186915887851, | |
| "grad_norm": 238278.5625, | |
| "learning_rate": 3.170731707317073e-05, | |
| "loss": 1.0462, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 5.257009345794392, | |
| "grad_norm": 270365.65625, | |
| "learning_rate": 3.094512195121951e-05, | |
| "loss": 1.041, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 5.373831775700935, | |
| "grad_norm": 258900.875, | |
| "learning_rate": 3.0182926829268294e-05, | |
| "loss": 1.0373, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 5.490654205607477, | |
| "grad_norm": 330339.34375, | |
| "learning_rate": 2.9420731707317074e-05, | |
| "loss": 1.0131, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 5.607476635514018, | |
| "grad_norm": 263223.84375, | |
| "learning_rate": 2.8658536585365854e-05, | |
| "loss": 1.0304, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 5.724299065420561, | |
| "grad_norm": 247210.890625, | |
| "learning_rate": 2.7896341463414637e-05, | |
| "loss": 1.0574, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 5.841121495327103, | |
| "grad_norm": 172526.59375, | |
| "learning_rate": 2.7134146341463417e-05, | |
| "loss": 1.026, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 5.957943925233645, | |
| "grad_norm": 229264.15625, | |
| "learning_rate": 2.6371951219512197e-05, | |
| "loss": 1.0295, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7354892205638475, | |
| "eval_loss": 0.9034654498100281, | |
| "eval_runtime": 15.3671, | |
| "eval_samples_per_second": 78.479, | |
| "eval_steps_per_second": 2.473, | |
| "step": 1284 | |
| }, | |
| { | |
| "epoch": 6.074766355140187, | |
| "grad_norm": 260367.5, | |
| "learning_rate": 2.5609756097560977e-05, | |
| "loss": 0.9993, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 6.191588785046729, | |
| "grad_norm": 318666.875, | |
| "learning_rate": 2.4847560975609756e-05, | |
| "loss": 0.9986, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 6.308411214953271, | |
| "grad_norm": 197853.28125, | |
| "learning_rate": 2.4085365853658536e-05, | |
| "loss": 1.0034, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 6.425233644859813, | |
| "grad_norm": 325670.03125, | |
| "learning_rate": 2.332317073170732e-05, | |
| "loss": 0.96, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 6.542056074766355, | |
| "grad_norm": 256457.5625, | |
| "learning_rate": 2.25609756097561e-05, | |
| "loss": 0.9791, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 6.658878504672897, | |
| "grad_norm": 318511.90625, | |
| "learning_rate": 2.179878048780488e-05, | |
| "loss": 0.9682, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 6.775700934579439, | |
| "grad_norm": 220621.96875, | |
| "learning_rate": 2.103658536585366e-05, | |
| "loss": 0.9827, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 6.892523364485982, | |
| "grad_norm": 253108.34375, | |
| "learning_rate": 2.0274390243902442e-05, | |
| "loss": 0.9602, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.75787728026534, | |
| "eval_loss": 0.8933297991752625, | |
| "eval_runtime": 15.7508, | |
| "eval_samples_per_second": 76.567, | |
| "eval_steps_per_second": 2.413, | |
| "step": 1498 | |
| }, | |
| { | |
| "epoch": 7.009345794392523, | |
| "grad_norm": 236817.09375, | |
| "learning_rate": 1.9512195121951222e-05, | |
| "loss": 0.9617, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 7.126168224299065, | |
| "grad_norm": 340729.375, | |
| "learning_rate": 1.8750000000000002e-05, | |
| "loss": 0.9256, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 7.242990654205608, | |
| "grad_norm": 229750.921875, | |
| "learning_rate": 1.798780487804878e-05, | |
| "loss": 0.9942, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 7.359813084112149, | |
| "grad_norm": 312539.53125, | |
| "learning_rate": 1.722560975609756e-05, | |
| "loss": 0.9413, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 7.4766355140186915, | |
| "grad_norm": 239815.234375, | |
| "learning_rate": 1.6463414634146345e-05, | |
| "loss": 0.9261, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 7.593457943925234, | |
| "grad_norm": 249483.65625, | |
| "learning_rate": 1.5701219512195124e-05, | |
| "loss": 0.9538, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 7.710280373831775, | |
| "grad_norm": 213290.171875, | |
| "learning_rate": 1.4939024390243902e-05, | |
| "loss": 0.9545, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 7.827102803738318, | |
| "grad_norm": 329573.375, | |
| "learning_rate": 1.4176829268292682e-05, | |
| "loss": 0.9201, | |
| "step": 1675 | |
| }, | |
| { | |
| "epoch": 7.94392523364486, | |
| "grad_norm": 302048.90625, | |
| "learning_rate": 1.3414634146341466e-05, | |
| "loss": 0.9428, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7678275290215588, | |
| "eval_loss": 0.8566215634346008, | |
| "eval_runtime": 15.2082, | |
| "eval_samples_per_second": 79.299, | |
| "eval_steps_per_second": 2.499, | |
| "step": 1712 | |
| }, | |
| { | |
| "epoch": 8.060747663551401, | |
| "grad_norm": 251469.65625, | |
| "learning_rate": 1.2652439024390245e-05, | |
| "loss": 0.9199, | |
| "step": 1725 | |
| }, | |
| { | |
| "epoch": 8.177570093457945, | |
| "grad_norm": 285863.75, | |
| "learning_rate": 1.1890243902439025e-05, | |
| "loss": 0.8714, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 8.294392523364486, | |
| "grad_norm": 307541.71875, | |
| "learning_rate": 1.1128048780487805e-05, | |
| "loss": 0.921, | |
| "step": 1775 | |
| }, | |
| { | |
| "epoch": 8.411214953271028, | |
| "grad_norm": 269017.46875, | |
| "learning_rate": 1.0365853658536585e-05, | |
| "loss": 0.8862, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 8.52803738317757, | |
| "grad_norm": 237760.375, | |
| "learning_rate": 9.603658536585366e-06, | |
| "loss": 0.924, | |
| "step": 1825 | |
| }, | |
| { | |
| "epoch": 8.644859813084112, | |
| "grad_norm": 289806.0625, | |
| "learning_rate": 8.841463414634146e-06, | |
| "loss": 0.9126, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 8.761682242990654, | |
| "grad_norm": 231987.5625, | |
| "learning_rate": 8.079268292682928e-06, | |
| "loss": 0.9011, | |
| "step": 1875 | |
| }, | |
| { | |
| "epoch": 8.878504672897197, | |
| "grad_norm": 319086.59375, | |
| "learning_rate": 7.317073170731707e-06, | |
| "loss": 0.8923, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 8.995327102803738, | |
| "grad_norm": 242763.078125, | |
| "learning_rate": 6.554878048780488e-06, | |
| "loss": 0.8893, | |
| "step": 1925 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.7611940298507462, | |
| "eval_loss": 0.8635094165802002, | |
| "eval_runtime": 15.4791, | |
| "eval_samples_per_second": 77.911, | |
| "eval_steps_per_second": 2.455, | |
| "step": 1926 | |
| }, | |
| { | |
| "epoch": 9.11214953271028, | |
| "grad_norm": 324516.875, | |
| "learning_rate": 5.792682926829269e-06, | |
| "loss": 0.8399, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 9.228971962616823, | |
| "grad_norm": 285448.65625, | |
| "learning_rate": 5.030487804878049e-06, | |
| "loss": 0.8979, | |
| "step": 1975 | |
| }, | |
| { | |
| "epoch": 9.345794392523365, | |
| "grad_norm": 266817.96875, | |
| "learning_rate": 4.26829268292683e-06, | |
| "loss": 0.8605, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 9.462616822429906, | |
| "grad_norm": 208118.15625, | |
| "learning_rate": 3.5060975609756102e-06, | |
| "loss": 0.8264, | |
| "step": 2025 | |
| }, | |
| { | |
| "epoch": 9.57943925233645, | |
| "grad_norm": 295613.59375, | |
| "learning_rate": 2.7439024390243905e-06, | |
| "loss": 0.9057, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 9.69626168224299, | |
| "grad_norm": 311793.40625, | |
| "learning_rate": 1.9817073170731707e-06, | |
| "loss": 0.9013, | |
| "step": 2075 | |
| }, | |
| { | |
| "epoch": 9.813084112149532, | |
| "grad_norm": 334048.90625, | |
| "learning_rate": 1.2195121951219514e-06, | |
| "loss": 0.8881, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 9.929906542056075, | |
| "grad_norm": 310285.15625, | |
| "learning_rate": 4.573170731707317e-07, | |
| "loss": 0.8661, | |
| "step": 2125 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7595356550580431, | |
| "eval_loss": 0.8694736361503601, | |
| "eval_runtime": 15.1828, | |
| "eval_samples_per_second": 79.432, | |
| "eval_steps_per_second": 2.503, | |
| "step": 2140 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 2140, | |
| "total_flos": 5.510586115727032e+19, | |
| "train_loss": 1.0744269477986843, | |
| "train_runtime": 1942.0913, | |
| "train_samples_per_second": 35.168, | |
| "train_steps_per_second": 1.102 | |
| } | |
| ], | |
| "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 | |
| } | |