| { | |
| "best_metric": 0.9788888888888889, | |
| "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": 7.109246253967285, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 2.2922, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 6.89300012588501, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 2.0928, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 6.730234622955322, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 1.6756, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 9.772377014160156, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 1.0287, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 17.15283966064453, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.5986, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 27.78217315673828, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.4796, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 11.25429630279541, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.4245, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 29.021835327148438, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.3994, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 15.250253677368164, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.3424, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 9.626118659973145, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.3437, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 14.916412353515625, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.3188, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 11.76590633392334, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.3324, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 8.164874076843262, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.3169, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 10.07582950592041, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.3342, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 7.93110466003418, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.2696, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 12.225959777832031, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.2655, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 9.060185432434082, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.2531, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 11.284368515014648, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.244, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 12.08557415008545, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2825, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.957037037037037, | |
| "eval_loss": 0.1273288130760193, | |
| "eval_runtime": 27.9178, | |
| "eval_samples_per_second": 96.713, | |
| "eval_steps_per_second": 3.045, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 9.616697311401367, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.2376, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 7.073214530944824, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.2612, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 9.791173934936523, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.2563, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 8.431086540222168, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.2412, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 8.49256706237793, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.2159, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 18.413179397583008, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.2413, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 10.551178932189941, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.2605, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 10.21691608428955, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.2021, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 5.306635856628418, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.2129, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 5.005842685699463, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.1824, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 8.03073501586914, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.197, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 9.381324768066406, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.2008, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 9.063702583312988, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.2149, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 7.741683006286621, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.1696, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 13.849589347839355, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.1918, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 9.223557472229004, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.173, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 6.596388339996338, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.1916, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 7.863681316375732, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.1772, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 15.154926300048828, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1404, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9733333333333334, | |
| "eval_loss": 0.07973578572273254, | |
| "eval_runtime": 32.4882, | |
| "eval_samples_per_second": 83.107, | |
| "eval_steps_per_second": 2.616, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 8.267267227172852, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.1721, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 6.031956195831299, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.1819, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 4.921173095703125, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.1735, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "grad_norm": 6.759762287139893, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.1523, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.26, | |
| "grad_norm": 8.071491241455078, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.1574, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 7.740592956542969, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.1413, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "grad_norm": 10.867220878601074, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.1445, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "grad_norm": 6.91710901260376, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.15, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.47, | |
| "grad_norm": 5.530294418334961, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.1222, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "grad_norm": 7.548226833343506, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.1442, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "grad_norm": 8.537381172180176, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.1485, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "grad_norm": 10.676304817199707, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.1551, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "grad_norm": 7.077390670776367, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.1716, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 2.74, | |
| "grad_norm": 8.090137481689453, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.1413, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 2.79, | |
| "grad_norm": 9.676934242248535, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.164, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "grad_norm": 18.995750427246094, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.1703, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "grad_norm": 5.193699359893799, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.1412, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 2.95, | |
| "grad_norm": 18.007177352905273, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.1192, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 15.536856651306152, | |
| "learning_rate": 0.0, | |
| "loss": 0.1394, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9788888888888889, | |
| "eval_loss": 0.061217330396175385, | |
| "eval_runtime": 43.6421, | |
| "eval_samples_per_second": 61.867, | |
| "eval_steps_per_second": 1.948, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 570, | |
| "total_flos": 1.8124066505760768e+18, | |
| "train_loss": 0.33600869262427613, | |
| "train_runtime": 3352.6448, | |
| "train_samples_per_second": 21.744, | |
| "train_steps_per_second": 0.17 | |
| } | |
| ], | |
| "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 | |
| } | |