| { | |
| "best_metric": 0.98, | |
| "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": 5.6030073165893555, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 2.3227, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 5.942467212677002, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 2.114, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 9.006362915039062, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.7361, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 14.675954818725586, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.1337, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 13.018393516540527, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.6849, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 29.24870491027832, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.4908, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 17.498170852661133, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.4472, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 30.422813415527344, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.4151, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 15.24944019317627, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.3611, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 18.759361267089844, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.3719, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 14.161396026611328, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.3395, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 9.216828346252441, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.3214, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 10.63085651397705, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.327, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 14.545869827270508, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.3075, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 14.948440551757812, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.2626, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 16.263952255249023, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.2731, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 8.593974113464355, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.2553, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 11.096720695495605, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.2696, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 9.849555015563965, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2994, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9603703703703703, | |
| "eval_loss": 0.12342998385429382, | |
| "eval_runtime": 17.2042, | |
| "eval_samples_per_second": 156.938, | |
| "eval_steps_per_second": 4.941, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 11.277251243591309, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.2387, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 8.723546028137207, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.2595, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 15.544955253601074, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.224, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 14.086812019348145, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.235, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 9.04621410369873, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.2281, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 11.80008602142334, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.2342, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 3.7134058475494385, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.222, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 6.554859638214111, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.2025, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 9.202827453613281, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.1814, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 8.696596145629883, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.203, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 10.952228546142578, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.1701, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 10.55245304107666, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.1732, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 12.811739921569824, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.1906, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 9.685540199279785, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.1975, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 8.672029495239258, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.1704, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 15.387055397033691, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.1845, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 9.411192893981934, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.1906, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 10.436761856079102, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.1715, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 12.051985740661621, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1853, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9740740740740741, | |
| "eval_loss": 0.07047658413648605, | |
| "eval_runtime": 16.9051, | |
| "eval_samples_per_second": 159.715, | |
| "eval_steps_per_second": 5.028, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 7.126999855041504, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.1961, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 9.795936584472656, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.1833, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 8.326896667480469, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1612, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 11.762768745422363, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1729, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 7.951204299926758, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.1818, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 8.159992218017578, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.1607, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 11.383317947387695, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.1588, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 4.525303363800049, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.1509, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 5.270012855529785, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.157, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 6.473080158233643, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.1543, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 15.744648933410645, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.1291, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 9.5807523727417, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.1292, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 5.534010887145996, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.1512, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 8.032577514648438, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.1347, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 24.584503173828125, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.1569, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 3.999915599822998, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.1664, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 9.104424476623535, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.1938, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 8.999449729919434, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.1519, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 11.300539016723633, | |
| "learning_rate": 0.0, | |
| "loss": 0.158, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.98, | |
| "eval_loss": 0.06010868772864342, | |
| "eval_runtime": 16.8705, | |
| "eval_samples_per_second": 160.043, | |
| "eval_steps_per_second": 5.038, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 570, | |
| "total_flos": 1.8124066505760768e+18, | |
| "train_loss": 0.34462205903571946, | |
| "train_runtime": 1109.7931, | |
| "train_samples_per_second": 65.688, | |
| "train_steps_per_second": 0.514 | |
| } | |
| ], | |
| "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 | |
| } | |