| { | |
| "best_metric": 0.9740740740740741, | |
| "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": 9.440265655517578, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 2.305, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 6.114067077636719, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 2.0991, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 8.871339797973633, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.6922, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 12.588781356811523, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.0473, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 32.035255432128906, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.6879, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 18.531063079833984, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.5213, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 14.835468292236328, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.4985, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 11.247536659240723, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.3847, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 13.409842491149902, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.3478, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 12.523353576660156, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.3296, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 16.897123336791992, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.3466, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 14.646390914916992, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.3356, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 14.801265716552734, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.2372, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 10.752660751342773, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.3163, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 8.766670227050781, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.2725, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 19.720884323120117, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.2708, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 15.061501502990723, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.2514, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 11.353556632995605, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.2595, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.882011413574219, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2723, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9633333333333334, | |
| "eval_loss": 0.11899825185537338, | |
| "eval_runtime": 31.0072, | |
| "eval_samples_per_second": 87.076, | |
| "eval_steps_per_second": 2.741, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 10.369582176208496, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.2288, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 11.584061622619629, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.252, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 9.439119338989258, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.2043, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 10.231059074401855, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.2835, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 10.476188659667969, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.2311, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 10.889532089233398, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.236, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 8.168774604797363, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.2129, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 11.831040382385254, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.2351, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 9.056485176086426, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.2328, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 6.355861663818359, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.2351, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 7.177761077880859, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.2055, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 9.945026397705078, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.1531, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 6.8051557540893555, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.1811, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 10.681591033935547, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.1561, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 12.849047660827637, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.1875, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 7.480422496795654, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.1912, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 10.37962818145752, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.179, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 5.792560577392578, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.1684, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 13.828116416931152, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1886, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9718518518518519, | |
| "eval_loss": 0.07636851072311401, | |
| "eval_runtime": 31.1849, | |
| "eval_samples_per_second": 86.58, | |
| "eval_steps_per_second": 2.726, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 9.011467933654785, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.2016, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 8.530644416809082, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.1711, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 10.602960586547852, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1852, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 6.965076923370361, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1609, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 12.922185897827148, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.153, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 11.024358749389648, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.1704, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 9.979962348937988, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.1789, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 12.759318351745605, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.1593, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 10.141521453857422, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.1692, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 8.66246223449707, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.1508, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 16.992340087890625, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.1788, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 5.300870418548584, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.1596, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 6.27362060546875, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.1566, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 7.903752326965332, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.1487, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 11.882786750793457, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.1951, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 9.480567932128906, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.152, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 8.694555282592773, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.1469, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 4.170577526092529, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.1475, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 10.84870433807373, | |
| "learning_rate": 0.0, | |
| "loss": 0.1336, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9740740740740741, | |
| "eval_loss": 0.06846870481967926, | |
| "eval_runtime": 31.5637, | |
| "eval_samples_per_second": 85.541, | |
| "eval_steps_per_second": 2.693, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 570, | |
| "total_flos": 1.8124066505760768e+18, | |
| "train_loss": 0.3430963271542599, | |
| "train_runtime": 1158.7002, | |
| "train_samples_per_second": 62.915, | |
| "train_steps_per_second": 0.492 | |
| } | |
| ], | |
| "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 | |
| } | |