| { | |
| "best_metric": 0.7697063369397218, | |
| "best_model_checkpoint": "resnet-152-fv-finetuned-memess/checkpoint-340", | |
| "epoch": 19.987654320987655, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 3e-05, | |
| "loss": 1.6137, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "learning_rate": 6e-05, | |
| "loss": 1.5902, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_accuracy": 0.4938176197836167, | |
| "eval_f1": 0.3529401365285427, | |
| "eval_loss": 1.5518949031829834, | |
| "eval_precision": 0.34914463899143217, | |
| "eval_recall": 0.4938176197836167, | |
| "eval_runtime": 12.637, | |
| "eval_samples_per_second": 102.398, | |
| "eval_steps_per_second": 1.662, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 9e-05, | |
| "loss": 1.5767, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "learning_rate": 0.00012, | |
| "loss": 1.4694, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.4891808346213292, | |
| "eval_f1": 0.3222387674010549, | |
| "eval_loss": 1.3730432987213135, | |
| "eval_precision": 0.409515151732489, | |
| "eval_recall": 0.4891808346213292, | |
| "eval_runtime": 11.6976, | |
| "eval_samples_per_second": 110.621, | |
| "eval_steps_per_second": 1.795, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.00011666666666666667, | |
| "loss": 1.4248, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 0.00011333333333333333, | |
| "loss": 1.3129, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.5301391035548686, | |
| "eval_f1": 0.4005333885311202, | |
| "eval_loss": 1.2052260637283325, | |
| "eval_precision": 0.35039701592103834, | |
| "eval_recall": 0.5301391035548686, | |
| "eval_runtime": 11.678, | |
| "eval_samples_per_second": 110.806, | |
| "eval_steps_per_second": 1.798, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.49, | |
| "learning_rate": 0.00011, | |
| "loss": 1.2909, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "learning_rate": 0.00010666666666666667, | |
| "loss": 1.1831, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.99, | |
| "eval_accuracy": 0.5587326120556414, | |
| "eval_f1": 0.44441640330937476, | |
| "eval_loss": 1.1141759157180786, | |
| "eval_precision": 0.4077366623668601, | |
| "eval_recall": 0.5587326120556414, | |
| "eval_runtime": 11.3977, | |
| "eval_samples_per_second": 113.532, | |
| "eval_steps_per_second": 1.842, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.49, | |
| "learning_rate": 0.00010333333333333334, | |
| "loss": 1.1723, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0581, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.99, | |
| "eval_accuracy": 0.6012364760432767, | |
| "eval_f1": 0.5107959944262404, | |
| "eval_loss": 0.9929684996604919, | |
| "eval_precision": 0.5680411145813762, | |
| "eval_recall": 0.6012364760432767, | |
| "eval_runtime": 11.68, | |
| "eval_samples_per_second": 110.787, | |
| "eval_steps_per_second": 1.798, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.49, | |
| "learning_rate": 9.666666666666667e-05, | |
| "loss": 1.0252, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "learning_rate": 9.333333333333334e-05, | |
| "loss": 0.9464, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.99, | |
| "eval_accuracy": 0.6506955177743431, | |
| "eval_f1": 0.6028776409086235, | |
| "eval_loss": 0.9262698888778687, | |
| "eval_precision": 0.6199847386276787, | |
| "eval_recall": 0.6506955177743431, | |
| "eval_runtime": 12.3363, | |
| "eval_samples_per_second": 104.894, | |
| "eval_steps_per_second": 1.702, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 9e-05, | |
| "loss": 0.9124, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "learning_rate": 8.666666666666667e-05, | |
| "loss": 0.8581, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.99, | |
| "eval_accuracy": 0.6916537867078826, | |
| "eval_f1": 0.6637680036287541, | |
| "eval_loss": 0.8399637341499329, | |
| "eval_precision": 0.6645266847727749, | |
| "eval_recall": 0.6916537867078826, | |
| "eval_runtime": 11.5743, | |
| "eval_samples_per_second": 111.8, | |
| "eval_steps_per_second": 1.814, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 7.49, | |
| "learning_rate": 8.333333333333333e-05, | |
| "loss": 0.7879, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 7.999999999999999e-05, | |
| "loss": 0.7739, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "eval_accuracy": 0.7086553323029366, | |
| "eval_f1": 0.6844873364202133, | |
| "eval_loss": 0.782887876033783, | |
| "eval_precision": 0.6918304772253453, | |
| "eval_recall": 0.7086553323029366, | |
| "eval_runtime": 11.6875, | |
| "eval_samples_per_second": 110.716, | |
| "eval_steps_per_second": 1.797, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.49, | |
| "learning_rate": 7.666666666666667e-05, | |
| "loss": 0.7217, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "learning_rate": 7.333333333333334e-05, | |
| "loss": 0.6762, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.99, | |
| "eval_accuracy": 0.731839258114374, | |
| "eval_f1": 0.7189272349665002, | |
| "eval_loss": 0.7512343525886536, | |
| "eval_precision": 0.720625431030895, | |
| "eval_recall": 0.731839258114374, | |
| "eval_runtime": 11.7148, | |
| "eval_samples_per_second": 110.459, | |
| "eval_steps_per_second": 1.793, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 9.49, | |
| "learning_rate": 7.000000000000001e-05, | |
| "loss": 0.629, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 0.6162, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.99, | |
| "eval_accuracy": 0.7264296754250387, | |
| "eval_f1": 0.7241444686897709, | |
| "eval_loss": 0.7409198880195618, | |
| "eval_precision": 0.7243707884059862, | |
| "eval_recall": 0.7264296754250387, | |
| "eval_runtime": 11.6853, | |
| "eval_samples_per_second": 110.738, | |
| "eval_steps_per_second": 1.797, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.49, | |
| "learning_rate": 6.333333333333333e-05, | |
| "loss": 0.5645, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "learning_rate": 6e-05, | |
| "loss": 0.5546, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.99, | |
| "eval_accuracy": 0.7465224111282844, | |
| "eval_f1": 0.739451747437607, | |
| "eval_loss": 0.6936307549476624, | |
| "eval_precision": 0.7428819169389939, | |
| "eval_recall": 0.7465224111282844, | |
| "eval_runtime": 12.631, | |
| "eval_samples_per_second": 102.446, | |
| "eval_steps_per_second": 1.663, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 11.49, | |
| "learning_rate": 5.6666666666666664e-05, | |
| "loss": 0.5366, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "learning_rate": 5.333333333333333e-05, | |
| "loss": 0.4633, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 11.99, | |
| "eval_accuracy": 0.7472952086553323, | |
| "eval_f1": 0.7412476847017665, | |
| "eval_loss": 0.6779463291168213, | |
| "eval_precision": 0.7392919538487581, | |
| "eval_recall": 0.7472952086553323, | |
| "eval_runtime": 11.6041, | |
| "eval_samples_per_second": 111.512, | |
| "eval_steps_per_second": 1.81, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.49, | |
| "learning_rate": 5e-05, | |
| "loss": 0.4926, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 0.4373, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.99, | |
| "eval_accuracy": 0.7573415765069552, | |
| "eval_f1": 0.7523125742209509, | |
| "eval_loss": 0.6735653281211853, | |
| "eval_precision": 0.7491811962248055, | |
| "eval_recall": 0.7573415765069552, | |
| "eval_runtime": 11.6552, | |
| "eval_samples_per_second": 111.023, | |
| "eval_steps_per_second": 1.802, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 13.49, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 0.4312, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "learning_rate": 3.9999999999999996e-05, | |
| "loss": 0.4074, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 13.99, | |
| "eval_accuracy": 0.7565687789799073, | |
| "eval_f1": 0.7527535377416857, | |
| "eval_loss": 0.6534111499786377, | |
| "eval_precision": 0.7516277057026929, | |
| "eval_recall": 0.7565687789799073, | |
| "eval_runtime": 11.4103, | |
| "eval_samples_per_second": 113.406, | |
| "eval_steps_per_second": 1.84, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.49, | |
| "learning_rate": 3.666666666666667e-05, | |
| "loss": 0.416, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.39, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.99, | |
| "eval_accuracy": 0.7650695517774343, | |
| "eval_f1": 0.7607628936527902, | |
| "eval_loss": 0.6520630121231079, | |
| "eval_precision": 0.7602577481853278, | |
| "eval_recall": 0.7650695517774343, | |
| "eval_runtime": 11.5842, | |
| "eval_samples_per_second": 111.704, | |
| "eval_steps_per_second": 1.813, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 15.49, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3763, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.3766, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 15.99, | |
| "eval_accuracy": 0.768160741885626, | |
| "eval_f1": 0.7630261314874124, | |
| "eval_loss": 0.6499382257461548, | |
| "eval_precision": 0.7606590123805609, | |
| "eval_recall": 0.768160741885626, | |
| "eval_runtime": 12.9158, | |
| "eval_samples_per_second": 100.187, | |
| "eval_steps_per_second": 1.626, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.49, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.3717, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "learning_rate": 1.9999999999999998e-05, | |
| "loss": 0.3507, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.99, | |
| "eval_accuracy": 0.7697063369397218, | |
| "eval_f1": 0.7686298015191362, | |
| "eval_loss": 0.6496546864509583, | |
| "eval_precision": 0.7686320213357392, | |
| "eval_recall": 0.7697063369397218, | |
| "eval_runtime": 11.8962, | |
| "eval_samples_per_second": 108.774, | |
| "eval_steps_per_second": 1.765, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 17.49, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.3484, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.3589, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 17.99, | |
| "eval_accuracy": 0.7534775888717156, | |
| "eval_f1": 0.7501567812782248, | |
| "eval_loss": 0.6518656611442566, | |
| "eval_precision": 0.7484569264769408, | |
| "eval_recall": 0.7534775888717156, | |
| "eval_runtime": 11.5335, | |
| "eval_samples_per_second": 112.195, | |
| "eval_steps_per_second": 1.821, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.49, | |
| "learning_rate": 9.999999999999999e-06, | |
| "loss": 0.3482, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.3261, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.99, | |
| "eval_accuracy": 0.758887171561051, | |
| "eval_f1": 0.7584538253621697, | |
| "eval_loss": 0.6448618769645691, | |
| "eval_precision": 0.7596824421701718, | |
| "eval_recall": 0.758887171561051, | |
| "eval_runtime": 11.4485, | |
| "eval_samples_per_second": 113.028, | |
| "eval_steps_per_second": 1.834, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 19.49, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.342, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "learning_rate": 0.0, | |
| "loss": 0.3234, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "eval_accuracy": 0.767387944358578, | |
| "eval_f1": 0.7646848616766787, | |
| "eval_loss": 0.6281499266624451, | |
| "eval_precision": 0.7651125602674349, | |
| "eval_recall": 0.767387944358578, | |
| "eval_runtime": 11.7608, | |
| "eval_samples_per_second": 110.027, | |
| "eval_steps_per_second": 1.786, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.99, | |
| "step": 400, | |
| "total_flos": 5.432231359979311e+18, | |
| "train_loss": 0.7463661253452301, | |
| "train_runtime": 1369.2901, | |
| "train_samples_per_second": 75.572, | |
| "train_steps_per_second": 0.292 | |
| } | |
| ], | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "total_flos": 5.432231359979311e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |