| { | |
| "best_metric": 0.5043315291404724, | |
| "best_model_checkpoint": "seizure_vit/seizure_vit_jlb_231112_fft_raw_combo/checkpoint-2000", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 8733, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 1.977098362532921e-06, | |
| "loss": 0.661, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.954196725065842e-06, | |
| "loss": 0.5647, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.931295087598763e-06, | |
| "loss": 0.5265, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.9083934501316844e-06, | |
| "loss": 0.5186, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.8854918126646053e-06, | |
| "loss": 0.4777, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_loss": 0.5236575603485107, | |
| "eval_roc_auc": 0.7454533473906167, | |
| "eval_runtime": 252.0313, | |
| "eval_samples_per_second": 60.215, | |
| "eval_steps_per_second": 7.527, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "learning_rate": 1.8625901751975266e-06, | |
| "loss": 0.4781, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.8396885377304477e-06, | |
| "loss": 0.4512, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.8167869002633686e-06, | |
| "loss": 0.4484, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.79388526279629e-06, | |
| "loss": 0.4431, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "learning_rate": 1.770983625329211e-06, | |
| "loss": 0.4469, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_loss": 0.5114009380340576, | |
| "eval_roc_auc": 0.7541512915129152, | |
| "eval_runtime": 254.0044, | |
| "eval_samples_per_second": 59.747, | |
| "eval_steps_per_second": 7.468, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.748081987862132e-06, | |
| "loss": 0.4332, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.7251803503950533e-06, | |
| "loss": 0.4353, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.7022787129279744e-06, | |
| "loss": 0.4137, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.6793770754608953e-06, | |
| "loss": 0.4306, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "learning_rate": 1.6564754379938164e-06, | |
| "loss": 0.4122, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "eval_loss": 0.5084041953086853, | |
| "eval_roc_auc": 0.7566552451238797, | |
| "eval_runtime": 251.639, | |
| "eval_samples_per_second": 60.309, | |
| "eval_steps_per_second": 7.539, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 1.6335738005267376e-06, | |
| "loss": 0.3944, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 1.6106721630596587e-06, | |
| "loss": 0.4001, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 1.5877705255925798e-06, | |
| "loss": 0.3981, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 1.564868888125501e-06, | |
| "loss": 0.3807, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 1.541967250658422e-06, | |
| "loss": 0.3904, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 0.5043315291404724, | |
| "eval_roc_auc": 0.7610701107011071, | |
| "eval_runtime": 253.2239, | |
| "eval_samples_per_second": 59.931, | |
| "eval_steps_per_second": 7.491, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 1.5190656131913431e-06, | |
| "loss": 0.3841, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 1.496163975724264e-06, | |
| "loss": 0.367, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 1.4732623382571854e-06, | |
| "loss": 0.3748, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 1.4503607007901065e-06, | |
| "loss": 0.3775, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "learning_rate": 1.4274590633230274e-06, | |
| "loss": 0.3619, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "eval_loss": 0.5282618999481201, | |
| "eval_roc_auc": 0.7608724301528729, | |
| "eval_runtime": 248.745, | |
| "eval_samples_per_second": 61.01, | |
| "eval_steps_per_second": 7.626, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 1.4045574258559487e-06, | |
| "loss": 0.3557, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 1.3816557883888698e-06, | |
| "loss": 0.3551, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 1.3587541509217907e-06, | |
| "loss": 0.3539, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.335852513454712e-06, | |
| "loss": 0.3686, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 1.3129508759876332e-06, | |
| "loss": 0.3528, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "eval_loss": 0.5351711511611938, | |
| "eval_roc_auc": 0.7517132314180285, | |
| "eval_runtime": 247.7427, | |
| "eval_samples_per_second": 61.257, | |
| "eval_steps_per_second": 7.657, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 1.290049238520554e-06, | |
| "loss": 0.3515, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 1.2671476010534752e-06, | |
| "loss": 0.3481, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1.2442459635863965e-06, | |
| "loss": 0.3509, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "learning_rate": 1.2213443261193174e-06, | |
| "loss": 0.3486, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 1.1984426886522385e-06, | |
| "loss": 0.3445, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 0.5338488221168518, | |
| "eval_roc_auc": 0.757248286768582, | |
| "eval_runtime": 258.0066, | |
| "eval_samples_per_second": 58.82, | |
| "eval_steps_per_second": 7.353, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 1.1755410511851599e-06, | |
| "loss": 0.3215, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 1.1526394137180808e-06, | |
| "loss": 0.3344, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 1.1297377762510019e-06, | |
| "loss": 0.3263, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 1.106836138783923e-06, | |
| "loss": 0.3255, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "learning_rate": 1.0839345013168441e-06, | |
| "loss": 0.3221, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "eval_loss": 0.5387703776359558, | |
| "eval_roc_auc": 0.7508566157090143, | |
| "eval_runtime": 249.4365, | |
| "eval_samples_per_second": 60.841, | |
| "eval_steps_per_second": 7.605, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "learning_rate": 1.0610328638497652e-06, | |
| "loss": 0.3388, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "learning_rate": 1.0381312263826861e-06, | |
| "loss": 0.3365, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "learning_rate": 1.0152295889156075e-06, | |
| "loss": 0.3256, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "learning_rate": 9.923279514485286e-07, | |
| "loss": 0.3057, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 9.694263139814495e-07, | |
| "loss": 0.3109, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "eval_loss": 0.5640743374824524, | |
| "eval_roc_auc": 0.7457828149710068, | |
| "eval_runtime": 248.3252, | |
| "eval_samples_per_second": 61.113, | |
| "eval_steps_per_second": 7.639, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 9.465246765143707e-07, | |
| "loss": 0.3301, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 9.236230390472919e-07, | |
| "loss": 0.309, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 9.007214015802129e-07, | |
| "loss": 0.3168, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "learning_rate": 8.778197641131341e-07, | |
| "loss": 0.3337, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "learning_rate": 8.549181266460551e-07, | |
| "loss": 0.3203, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "eval_loss": 0.5403934717178345, | |
| "eval_roc_auc": 0.757445967316816, | |
| "eval_runtime": 249.3451, | |
| "eval_samples_per_second": 60.863, | |
| "eval_steps_per_second": 7.608, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 8.320164891789763e-07, | |
| "loss": 0.3144, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "learning_rate": 8.091148517118974e-07, | |
| "loss": 0.3166, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 7.862132142448184e-07, | |
| "loss": 0.3091, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "learning_rate": 7.633115767777396e-07, | |
| "loss": 0.3097, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 7.404099393106607e-07, | |
| "loss": 0.294, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_loss": 0.5420622229576111, | |
| "eval_roc_auc": 0.7563916710595677, | |
| "eval_runtime": 247.7049, | |
| "eval_samples_per_second": 61.266, | |
| "eval_steps_per_second": 7.658, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 7.175083018435818e-07, | |
| "loss": 0.2897, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 6.946066643765029e-07, | |
| "loss": 0.3126, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 6.71705026909424e-07, | |
| "loss": 0.2998, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "learning_rate": 6.488033894423451e-07, | |
| "loss": 0.2941, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 6.259017519752662e-07, | |
| "loss": 0.2964, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "eval_loss": 0.5581579208374023, | |
| "eval_roc_auc": 0.7493410648392198, | |
| "eval_runtime": 253.738, | |
| "eval_samples_per_second": 59.81, | |
| "eval_steps_per_second": 7.476, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 6.030001145081873e-07, | |
| "loss": 0.296, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 5.800984770411084e-07, | |
| "loss": 0.3026, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "learning_rate": 5.571968395740295e-07, | |
| "loss": 0.2958, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "learning_rate": 5.342952021069507e-07, | |
| "loss": 0.2778, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 5.113935646398717e-07, | |
| "loss": 0.292, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "eval_loss": 0.5513432621955872, | |
| "eval_roc_auc": 0.7560622034791777, | |
| "eval_runtime": 244.5312, | |
| "eval_samples_per_second": 62.062, | |
| "eval_steps_per_second": 7.758, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 4.884919271727928e-07, | |
| "loss": 0.294, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "learning_rate": 4.6559028970571394e-07, | |
| "loss": 0.295, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 2.34, | |
| "learning_rate": 4.4268865223863505e-07, | |
| "loss": 0.2914, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 4.1978701477155617e-07, | |
| "loss": 0.2771, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "learning_rate": 3.9688537730447723e-07, | |
| "loss": 0.2838, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "eval_loss": 0.5557020306587219, | |
| "eval_roc_auc": 0.7597522403795467, | |
| "eval_runtime": 252.0114, | |
| "eval_samples_per_second": 60.219, | |
| "eval_steps_per_second": 7.527, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 3.7398373983739835e-07, | |
| "loss": 0.3156, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "learning_rate": 3.5108210237031946e-07, | |
| "loss": 0.2875, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 2.51, | |
| "learning_rate": 3.281804649032406e-07, | |
| "loss": 0.2779, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 3.052788274361617e-07, | |
| "loss": 0.2959, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 2.8237718996908276e-07, | |
| "loss": 0.2736, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "eval_loss": 0.5514435172080994, | |
| "eval_roc_auc": 0.760608856088561, | |
| "eval_runtime": 250.5016, | |
| "eval_samples_per_second": 60.582, | |
| "eval_steps_per_second": 7.573, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 2.5947555250200387e-07, | |
| "loss": 0.2989, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 2.36573915034925e-07, | |
| "loss": 0.2889, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 2.1367227756784608e-07, | |
| "loss": 0.2814, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 2.71, | |
| "learning_rate": 1.907706401007672e-07, | |
| "loss": 0.2917, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "learning_rate": 1.678690026336883e-07, | |
| "loss": 0.2922, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.75, | |
| "eval_loss": 0.550301730632782, | |
| "eval_roc_auc": 0.7537559304164471, | |
| "eval_runtime": 244.8445, | |
| "eval_samples_per_second": 61.982, | |
| "eval_steps_per_second": 7.748, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 2.78, | |
| "learning_rate": 1.449673651666094e-07, | |
| "loss": 0.2588, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 1.2206572769953052e-07, | |
| "loss": 0.2947, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 9.916409023245162e-08, | |
| "loss": 0.2901, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 7.626245276537272e-08, | |
| "loss": 0.2827, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 5.3360815298293824e-08, | |
| "loss": 0.2699, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "eval_loss": 0.553462564945221, | |
| "eval_roc_auc": 0.757841328413284, | |
| "eval_runtime": 246.3262, | |
| "eval_samples_per_second": 61.609, | |
| "eval_steps_per_second": 7.701, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "learning_rate": 3.0459177831214926e-08, | |
| "loss": 0.2763, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 7.557540364136036e-09, | |
| "loss": 0.2881, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 8733, | |
| "total_flos": 2.165002202648317e+19, | |
| "train_loss": 0.34656610874380017, | |
| "train_runtime": 7351.286, | |
| "train_samples_per_second": 38.005, | |
| "train_steps_per_second": 1.188 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 8733, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.165002202648317e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |