|
{ |
|
"best_metric": 0.9777777777777777, |
|
"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": 6.905220031738281, |
|
"learning_rate": 8.771929824561403e-06, |
|
"loss": 2.2755, |
|
"step": 10 |
|
}, |
|
{ |
|
"epoch": 0.11, |
|
"grad_norm": 5.883072853088379, |
|
"learning_rate": 1.7543859649122806e-05, |
|
"loss": 2.0818, |
|
"step": 20 |
|
}, |
|
{ |
|
"epoch": 0.16, |
|
"grad_norm": 6.574605941772461, |
|
"learning_rate": 2.6315789473684212e-05, |
|
"loss": 1.6674, |
|
"step": 30 |
|
}, |
|
{ |
|
"epoch": 0.21, |
|
"grad_norm": 18.086021423339844, |
|
"learning_rate": 3.508771929824561e-05, |
|
"loss": 1.0708, |
|
"step": 40 |
|
}, |
|
{ |
|
"epoch": 0.26, |
|
"grad_norm": 13.633099555969238, |
|
"learning_rate": 4.3859649122807014e-05, |
|
"loss": 0.6766, |
|
"step": 50 |
|
}, |
|
{ |
|
"epoch": 0.32, |
|
"grad_norm": 10.846696853637695, |
|
"learning_rate": 4.970760233918128e-05, |
|
"loss": 0.5076, |
|
"step": 60 |
|
}, |
|
{ |
|
"epoch": 0.37, |
|
"grad_norm": 18.08544921875, |
|
"learning_rate": 4.8732943469785574e-05, |
|
"loss": 0.4044, |
|
"step": 70 |
|
}, |
|
{ |
|
"epoch": 0.42, |
|
"grad_norm": 18.805522918701172, |
|
"learning_rate": 4.7758284600389865e-05, |
|
"loss": 0.4289, |
|
"step": 80 |
|
}, |
|
{ |
|
"epoch": 0.47, |
|
"grad_norm": 25.817827224731445, |
|
"learning_rate": 4.678362573099415e-05, |
|
"loss": 0.3679, |
|
"step": 90 |
|
}, |
|
{ |
|
"epoch": 0.53, |
|
"grad_norm": 19.707490921020508, |
|
"learning_rate": 4.580896686159844e-05, |
|
"loss": 0.3635, |
|
"step": 100 |
|
}, |
|
{ |
|
"epoch": 0.58, |
|
"grad_norm": 12.682143211364746, |
|
"learning_rate": 4.483430799220273e-05, |
|
"loss": 0.3333, |
|
"step": 110 |
|
}, |
|
{ |
|
"epoch": 0.63, |
|
"grad_norm": 10.674261093139648, |
|
"learning_rate": 4.3859649122807014e-05, |
|
"loss": 0.2928, |
|
"step": 120 |
|
}, |
|
{ |
|
"epoch": 0.68, |
|
"grad_norm": 11.278220176696777, |
|
"learning_rate": 4.2884990253411305e-05, |
|
"loss": 0.2902, |
|
"step": 130 |
|
}, |
|
{ |
|
"epoch": 0.74, |
|
"grad_norm": 13.635940551757812, |
|
"learning_rate": 4.1910331384015596e-05, |
|
"loss": 0.2921, |
|
"step": 140 |
|
}, |
|
{ |
|
"epoch": 0.79, |
|
"grad_norm": 9.778814315795898, |
|
"learning_rate": 4.093567251461988e-05, |
|
"loss": 0.2716, |
|
"step": 150 |
|
}, |
|
{ |
|
"epoch": 0.84, |
|
"grad_norm": 25.61815643310547, |
|
"learning_rate": 3.996101364522417e-05, |
|
"loss": 0.2775, |
|
"step": 160 |
|
}, |
|
{ |
|
"epoch": 0.89, |
|
"grad_norm": 13.44371223449707, |
|
"learning_rate": 3.898635477582846e-05, |
|
"loss": 0.2516, |
|
"step": 170 |
|
}, |
|
{ |
|
"epoch": 0.95, |
|
"grad_norm": 9.713900566101074, |
|
"learning_rate": 3.8011695906432746e-05, |
|
"loss": 0.2305, |
|
"step": 180 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"grad_norm": 10.197415351867676, |
|
"learning_rate": 3.7037037037037037e-05, |
|
"loss": 0.2542, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 1.0, |
|
"eval_accuracy": 0.9614814814814815, |
|
"eval_loss": 0.12235087901353836, |
|
"eval_runtime": 30.8807, |
|
"eval_samples_per_second": 87.433, |
|
"eval_steps_per_second": 2.753, |
|
"step": 190 |
|
}, |
|
{ |
|
"epoch": 1.05, |
|
"grad_norm": 9.8814697265625, |
|
"learning_rate": 3.606237816764133e-05, |
|
"loss": 0.2462, |
|
"step": 200 |
|
}, |
|
{ |
|
"epoch": 1.11, |
|
"grad_norm": 11.23945140838623, |
|
"learning_rate": 3.508771929824561e-05, |
|
"loss": 0.2386, |
|
"step": 210 |
|
}, |
|
{ |
|
"epoch": 1.16, |
|
"grad_norm": 11.149582862854004, |
|
"learning_rate": 3.41130604288499e-05, |
|
"loss": 0.2409, |
|
"step": 220 |
|
}, |
|
{ |
|
"epoch": 1.21, |
|
"grad_norm": 12.31209659576416, |
|
"learning_rate": 3.313840155945419e-05, |
|
"loss": 0.2301, |
|
"step": 230 |
|
}, |
|
{ |
|
"epoch": 1.26, |
|
"grad_norm": 28.526798248291016, |
|
"learning_rate": 3.216374269005848e-05, |
|
"loss": 0.2241, |
|
"step": 240 |
|
}, |
|
{ |
|
"epoch": 1.32, |
|
"grad_norm": 12.132701873779297, |
|
"learning_rate": 3.118908382066277e-05, |
|
"loss": 0.2201, |
|
"step": 250 |
|
}, |
|
{ |
|
"epoch": 1.37, |
|
"grad_norm": 13.534875869750977, |
|
"learning_rate": 3.0214424951267055e-05, |
|
"loss": 0.2013, |
|
"step": 260 |
|
}, |
|
{ |
|
"epoch": 1.42, |
|
"grad_norm": 9.8684720993042, |
|
"learning_rate": 2.9239766081871346e-05, |
|
"loss": 0.2075, |
|
"step": 270 |
|
}, |
|
{ |
|
"epoch": 1.47, |
|
"grad_norm": 17.482439041137695, |
|
"learning_rate": 2.8265107212475634e-05, |
|
"loss": 0.1876, |
|
"step": 280 |
|
}, |
|
{ |
|
"epoch": 1.53, |
|
"grad_norm": 6.973119258880615, |
|
"learning_rate": 2.729044834307992e-05, |
|
"loss": 0.1875, |
|
"step": 290 |
|
}, |
|
{ |
|
"epoch": 1.58, |
|
"grad_norm": 7.083260536193848, |
|
"learning_rate": 2.6315789473684212e-05, |
|
"loss": 0.181, |
|
"step": 300 |
|
}, |
|
{ |
|
"epoch": 1.63, |
|
"grad_norm": 9.20019245147705, |
|
"learning_rate": 2.53411306042885e-05, |
|
"loss": 0.1562, |
|
"step": 310 |
|
}, |
|
{ |
|
"epoch": 1.68, |
|
"grad_norm": 14.853752136230469, |
|
"learning_rate": 2.4366471734892787e-05, |
|
"loss": 0.1644, |
|
"step": 320 |
|
}, |
|
{ |
|
"epoch": 1.74, |
|
"grad_norm": 6.650509357452393, |
|
"learning_rate": 2.3391812865497074e-05, |
|
"loss": 0.1794, |
|
"step": 330 |
|
}, |
|
{ |
|
"epoch": 1.79, |
|
"grad_norm": 7.536526679992676, |
|
"learning_rate": 2.2417153996101365e-05, |
|
"loss": 0.1369, |
|
"step": 340 |
|
}, |
|
{ |
|
"epoch": 1.84, |
|
"grad_norm": 9.41813850402832, |
|
"learning_rate": 2.1442495126705653e-05, |
|
"loss": 0.199, |
|
"step": 350 |
|
}, |
|
{ |
|
"epoch": 1.89, |
|
"grad_norm": 10.232301712036133, |
|
"learning_rate": 2.046783625730994e-05, |
|
"loss": 0.1678, |
|
"step": 360 |
|
}, |
|
{ |
|
"epoch": 1.95, |
|
"grad_norm": 8.716858863830566, |
|
"learning_rate": 1.949317738791423e-05, |
|
"loss": 0.1889, |
|
"step": 370 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"grad_norm": 9.904465675354004, |
|
"learning_rate": 1.8518518518518518e-05, |
|
"loss": 0.171, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 2.0, |
|
"eval_accuracy": 0.9688888888888889, |
|
"eval_loss": 0.08309615403413773, |
|
"eval_runtime": 31.5481, |
|
"eval_samples_per_second": 85.584, |
|
"eval_steps_per_second": 2.694, |
|
"step": 380 |
|
}, |
|
{ |
|
"epoch": 2.05, |
|
"grad_norm": 11.40892505645752, |
|
"learning_rate": 1.7543859649122806e-05, |
|
"loss": 0.1923, |
|
"step": 390 |
|
}, |
|
{ |
|
"epoch": 2.11, |
|
"grad_norm": 15.750476837158203, |
|
"learning_rate": 1.6569200779727097e-05, |
|
"loss": 0.1873, |
|
"step": 400 |
|
}, |
|
{ |
|
"epoch": 2.16, |
|
"grad_norm": 10.811800003051758, |
|
"learning_rate": 1.5594541910331384e-05, |
|
"loss": 0.1703, |
|
"step": 410 |
|
}, |
|
{ |
|
"epoch": 2.21, |
|
"grad_norm": 16.07097816467285, |
|
"learning_rate": 1.4619883040935673e-05, |
|
"loss": 0.1624, |
|
"step": 420 |
|
}, |
|
{ |
|
"epoch": 2.26, |
|
"grad_norm": 10.764117240905762, |
|
"learning_rate": 1.364522417153996e-05, |
|
"loss": 0.1532, |
|
"step": 430 |
|
}, |
|
{ |
|
"epoch": 2.32, |
|
"grad_norm": 3.9213733673095703, |
|
"learning_rate": 1.267056530214425e-05, |
|
"loss": 0.1373, |
|
"step": 440 |
|
}, |
|
{ |
|
"epoch": 2.37, |
|
"grad_norm": 6.07966947555542, |
|
"learning_rate": 1.1695906432748537e-05, |
|
"loss": 0.16, |
|
"step": 450 |
|
}, |
|
{ |
|
"epoch": 2.42, |
|
"grad_norm": 13.090170860290527, |
|
"learning_rate": 1.0721247563352826e-05, |
|
"loss": 0.1317, |
|
"step": 460 |
|
}, |
|
{ |
|
"epoch": 2.47, |
|
"grad_norm": 9.082914352416992, |
|
"learning_rate": 9.746588693957115e-06, |
|
"loss": 0.14, |
|
"step": 470 |
|
}, |
|
{ |
|
"epoch": 2.53, |
|
"grad_norm": 7.2037248611450195, |
|
"learning_rate": 8.771929824561403e-06, |
|
"loss": 0.1731, |
|
"step": 480 |
|
}, |
|
{ |
|
"epoch": 2.58, |
|
"grad_norm": 6.6279120445251465, |
|
"learning_rate": 7.797270955165692e-06, |
|
"loss": 0.1256, |
|
"step": 490 |
|
}, |
|
{ |
|
"epoch": 2.63, |
|
"grad_norm": 7.699619770050049, |
|
"learning_rate": 6.82261208576998e-06, |
|
"loss": 0.12, |
|
"step": 500 |
|
}, |
|
{ |
|
"epoch": 2.68, |
|
"grad_norm": 7.525733947753906, |
|
"learning_rate": 5.8479532163742686e-06, |
|
"loss": 0.1514, |
|
"step": 510 |
|
}, |
|
{ |
|
"epoch": 2.74, |
|
"grad_norm": 6.572248458862305, |
|
"learning_rate": 4.873294346978558e-06, |
|
"loss": 0.1316, |
|
"step": 520 |
|
}, |
|
{ |
|
"epoch": 2.79, |
|
"grad_norm": 17.000349044799805, |
|
"learning_rate": 3.898635477582846e-06, |
|
"loss": 0.1674, |
|
"step": 530 |
|
}, |
|
{ |
|
"epoch": 2.84, |
|
"grad_norm": 8.921578407287598, |
|
"learning_rate": 2.9239766081871343e-06, |
|
"loss": 0.1624, |
|
"step": 540 |
|
}, |
|
{ |
|
"epoch": 2.89, |
|
"grad_norm": 5.688661098480225, |
|
"learning_rate": 1.949317738791423e-06, |
|
"loss": 0.1641, |
|
"step": 550 |
|
}, |
|
{ |
|
"epoch": 2.95, |
|
"grad_norm": 7.114222526550293, |
|
"learning_rate": 9.746588693957115e-07, |
|
"loss": 0.152, |
|
"step": 560 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"grad_norm": 9.845104217529297, |
|
"learning_rate": 0.0, |
|
"loss": 0.1617, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"eval_accuracy": 0.9777777777777777, |
|
"eval_loss": 0.06170220300555229, |
|
"eval_runtime": 31.1379, |
|
"eval_samples_per_second": 86.711, |
|
"eval_steps_per_second": 2.73, |
|
"step": 570 |
|
}, |
|
{ |
|
"epoch": 3.0, |
|
"step": 570, |
|
"total_flos": 1.8124066505760768e+18, |
|
"train_loss": 0.33351591700001765, |
|
"train_runtime": 1156.1276, |
|
"train_samples_per_second": 63.055, |
|
"train_steps_per_second": 0.493 |
|
} |
|
], |
|
"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 |
|
} |
|
|