| { | |
| "best_metric": 0.9814814814814815, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-570", | |
| "epoch": 3.0, | |
| "eval_steps": 500, | |
| "global_step": 570, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 5.7690863609313965, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 2.2623, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 7.180535316467285, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 2.0467, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 12.502141952514648, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.6134, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 12.182085037231445, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.0089, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 15.876863479614258, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.6195, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 23.71845054626465, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.5156, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 15.944512367248535, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.4736, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 22.624839782714844, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.3892, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 13.724577903747559, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.3564, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 9.811408996582031, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.3488, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 7.73867654800415, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.3028, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 14.532332420349121, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.2982, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 17.203603744506836, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.2659, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 17.189725875854492, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.2745, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 12.759488105773926, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.2586, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 15.634223937988281, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.2208, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 17.382017135620117, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.2191, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 23.635671615600586, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.2517, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 17.320865631103516, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2747, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9562962962962963, | |
| "eval_loss": 0.13819058239459991, | |
| "eval_runtime": 17.3473, | |
| "eval_samples_per_second": 155.644, | |
| "eval_steps_per_second": 4.9, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 6.783141613006592, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.1881, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 12.496849060058594, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.2501, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 11.41723346710205, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.2282, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 12.789288520812988, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.2056, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 15.091114044189453, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.1959, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 13.9872465133667, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.192, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 12.632612228393555, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.2159, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 5.643543243408203, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.2369, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 12.24279499053955, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.2166, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 9.774775505065918, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.1558, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 11.163515090942383, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.1835, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 15.248787879943848, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.2193, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 13.8126220703125, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.1867, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 5.466386795043945, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.1383, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 11.421280860900879, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.1611, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 3.8385016918182373, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.1518, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 15.16893196105957, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.198, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 8.695472717285156, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.1843, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 10.149725914001465, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1928, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9714814814814815, | |
| "eval_loss": 0.09133487939834595, | |
| "eval_runtime": 17.3865, | |
| "eval_samples_per_second": 155.293, | |
| "eval_steps_per_second": 4.889, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 8.00893783569336, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.1681, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 16.080177307128906, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.1489, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 13.361900329589844, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1362, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 11.800583839416504, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1489, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 13.107502937316895, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.1813, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 8.709113121032715, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.1874, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 6.434211730957031, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.1782, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 18.821304321289062, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.1365, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 5.870109558105469, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.1518, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 3.519899845123291, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.139, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 23.94843292236328, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.1486, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 5.265621185302734, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.1329, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 4.568530559539795, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.1513, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 8.631195068359375, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.1549, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 6.977248191833496, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.1465, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 12.340450286865234, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.1662, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 6.968770980834961, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.1449, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 6.5286688804626465, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.1513, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 9.231646537780762, | |
| "learning_rate": 0.0, | |
| "loss": 0.1317, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9814814814814815, | |
| "eval_loss": 0.06550158560276031, | |
| "eval_runtime": 17.1307, | |
| "eval_samples_per_second": 157.612, | |
| "eval_steps_per_second": 4.962, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 570, | |
| "total_flos": 1.8124066505760768e+18, | |
| "train_loss": 0.3264210811832495, | |
| "train_runtime": 1089.539, | |
| "train_samples_per_second": 66.909, | |
| "train_steps_per_second": 0.523 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 570, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 1.8124066505760768e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |