andrecastro's picture
Training in progress, epoch 1
e09a725
raw
history blame
15.6 kB
{
"best_metric": 0.9896390374331551,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-748",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 1122,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 4.424778761061947e-06,
"loss": 0.0312,
"step": 10
},
{
"epoch": 0.05,
"learning_rate": 8.849557522123894e-06,
"loss": 0.0446,
"step": 20
},
{
"epoch": 0.08,
"learning_rate": 1.3274336283185843e-05,
"loss": 0.0453,
"step": 30
},
{
"epoch": 0.11,
"learning_rate": 1.7699115044247787e-05,
"loss": 0.0327,
"step": 40
},
{
"epoch": 0.13,
"learning_rate": 2.2123893805309738e-05,
"loss": 0.1043,
"step": 50
},
{
"epoch": 0.16,
"learning_rate": 2.6548672566371686e-05,
"loss": 0.0937,
"step": 60
},
{
"epoch": 0.19,
"learning_rate": 3.097345132743363e-05,
"loss": 0.0594,
"step": 70
},
{
"epoch": 0.21,
"learning_rate": 3.5398230088495574e-05,
"loss": 0.0572,
"step": 80
},
{
"epoch": 0.24,
"learning_rate": 3.982300884955752e-05,
"loss": 0.1041,
"step": 90
},
{
"epoch": 0.27,
"learning_rate": 4.4247787610619477e-05,
"loss": 0.0644,
"step": 100
},
{
"epoch": 0.29,
"learning_rate": 4.867256637168142e-05,
"loss": 0.0606,
"step": 110
},
{
"epoch": 0.32,
"learning_rate": 4.9653121902874134e-05,
"loss": 0.0921,
"step": 120
},
{
"epoch": 0.35,
"learning_rate": 4.9157581764122896e-05,
"loss": 0.0753,
"step": 130
},
{
"epoch": 0.37,
"learning_rate": 4.866204162537166e-05,
"loss": 0.1348,
"step": 140
},
{
"epoch": 0.4,
"learning_rate": 4.816650148662042e-05,
"loss": 0.0963,
"step": 150
},
{
"epoch": 0.43,
"learning_rate": 4.767096134786918e-05,
"loss": 0.1455,
"step": 160
},
{
"epoch": 0.45,
"learning_rate": 4.717542120911794e-05,
"loss": 0.087,
"step": 170
},
{
"epoch": 0.48,
"learning_rate": 4.66798810703667e-05,
"loss": 0.0817,
"step": 180
},
{
"epoch": 0.51,
"learning_rate": 4.618434093161546e-05,
"loss": 0.0676,
"step": 190
},
{
"epoch": 0.53,
"learning_rate": 4.568880079286422e-05,
"loss": 0.1201,
"step": 200
},
{
"epoch": 0.56,
"learning_rate": 4.5193260654112984e-05,
"loss": 0.0388,
"step": 210
},
{
"epoch": 0.59,
"learning_rate": 4.4697720515361746e-05,
"loss": 0.1174,
"step": 220
},
{
"epoch": 0.61,
"learning_rate": 4.420218037661051e-05,
"loss": 0.049,
"step": 230
},
{
"epoch": 0.64,
"learning_rate": 4.370664023785927e-05,
"loss": 0.0693,
"step": 240
},
{
"epoch": 0.67,
"learning_rate": 4.321110009910803e-05,
"loss": 0.1011,
"step": 250
},
{
"epoch": 0.7,
"learning_rate": 4.27155599603568e-05,
"loss": 0.0721,
"step": 260
},
{
"epoch": 0.72,
"learning_rate": 4.222001982160555e-05,
"loss": 0.0908,
"step": 270
},
{
"epoch": 0.75,
"learning_rate": 4.172447968285431e-05,
"loss": 0.0502,
"step": 280
},
{
"epoch": 0.78,
"learning_rate": 4.122893954410307e-05,
"loss": 0.2335,
"step": 290
},
{
"epoch": 0.8,
"learning_rate": 4.0733399405351834e-05,
"loss": 0.0426,
"step": 300
},
{
"epoch": 0.83,
"learning_rate": 4.0237859266600596e-05,
"loss": 0.0492,
"step": 310
},
{
"epoch": 0.86,
"learning_rate": 3.9742319127849357e-05,
"loss": 0.0741,
"step": 320
},
{
"epoch": 0.88,
"learning_rate": 3.924677898909812e-05,
"loss": 0.1172,
"step": 330
},
{
"epoch": 0.91,
"learning_rate": 3.8751238850346886e-05,
"loss": 0.0826,
"step": 340
},
{
"epoch": 0.94,
"learning_rate": 3.825569871159564e-05,
"loss": 0.1365,
"step": 350
},
{
"epoch": 0.96,
"learning_rate": 3.77601585728444e-05,
"loss": 0.0935,
"step": 360
},
{
"epoch": 0.99,
"learning_rate": 3.726461843409316e-05,
"loss": 0.1073,
"step": 370
},
{
"epoch": 1.0,
"eval_accuracy": 0.9886363636363636,
"eval_confusion_matrix": [
[
1502,
10
],
[
24,
1456
]
],
"eval_loss": 0.03868470713496208,
"eval_precision": 0.9886780366084972,
"eval_recall": 0.9886363636363636,
"eval_runtime": 58.4232,
"eval_samples_per_second": 51.213,
"eval_steps_per_second": 6.402,
"step": 374
},
{
"epoch": 1.02,
"learning_rate": 3.676907829534192e-05,
"loss": 0.0938,
"step": 380
},
{
"epoch": 1.04,
"learning_rate": 3.6273538156590684e-05,
"loss": 0.0826,
"step": 390
},
{
"epoch": 1.07,
"learning_rate": 3.5777998017839445e-05,
"loss": 0.1043,
"step": 400
},
{
"epoch": 1.1,
"learning_rate": 3.5282457879088207e-05,
"loss": 0.0778,
"step": 410
},
{
"epoch": 1.12,
"learning_rate": 3.4786917740336974e-05,
"loss": 0.1392,
"step": 420
},
{
"epoch": 1.15,
"learning_rate": 3.429137760158573e-05,
"loss": 0.0918,
"step": 430
},
{
"epoch": 1.18,
"learning_rate": 3.379583746283449e-05,
"loss": 0.0726,
"step": 440
},
{
"epoch": 1.2,
"learning_rate": 3.330029732408325e-05,
"loss": 0.0928,
"step": 450
},
{
"epoch": 1.23,
"learning_rate": 3.280475718533201e-05,
"loss": 0.036,
"step": 460
},
{
"epoch": 1.26,
"learning_rate": 3.230921704658077e-05,
"loss": 0.0474,
"step": 470
},
{
"epoch": 1.28,
"learning_rate": 3.1813676907829534e-05,
"loss": 0.0968,
"step": 480
},
{
"epoch": 1.31,
"learning_rate": 3.13181367690783e-05,
"loss": 0.0828,
"step": 490
},
{
"epoch": 1.34,
"learning_rate": 3.0822596630327057e-05,
"loss": 0.0389,
"step": 500
},
{
"epoch": 1.36,
"learning_rate": 3.0327056491575818e-05,
"loss": 0.0749,
"step": 510
},
{
"epoch": 1.39,
"learning_rate": 2.983151635282458e-05,
"loss": 0.0756,
"step": 520
},
{
"epoch": 1.42,
"learning_rate": 2.933597621407334e-05,
"loss": 0.06,
"step": 530
},
{
"epoch": 1.44,
"learning_rate": 2.88404360753221e-05,
"loss": 0.0754,
"step": 540
},
{
"epoch": 1.47,
"learning_rate": 2.8344895936570865e-05,
"loss": 0.0667,
"step": 550
},
{
"epoch": 1.5,
"learning_rate": 2.7849355797819627e-05,
"loss": 0.1152,
"step": 560
},
{
"epoch": 1.52,
"learning_rate": 2.7353815659068388e-05,
"loss": 0.0889,
"step": 570
},
{
"epoch": 1.55,
"learning_rate": 2.6858275520317145e-05,
"loss": 0.107,
"step": 580
},
{
"epoch": 1.58,
"learning_rate": 2.6362735381565906e-05,
"loss": 0.0719,
"step": 590
},
{
"epoch": 1.6,
"learning_rate": 2.5867195242814668e-05,
"loss": 0.0487,
"step": 600
},
{
"epoch": 1.63,
"learning_rate": 2.537165510406343e-05,
"loss": 0.0611,
"step": 610
},
{
"epoch": 1.66,
"learning_rate": 2.4876114965312193e-05,
"loss": 0.0819,
"step": 620
},
{
"epoch": 1.68,
"learning_rate": 2.4380574826560954e-05,
"loss": 0.0512,
"step": 630
},
{
"epoch": 1.71,
"learning_rate": 2.3885034687809712e-05,
"loss": 0.0609,
"step": 640
},
{
"epoch": 1.74,
"learning_rate": 2.3389494549058473e-05,
"loss": 0.0586,
"step": 650
},
{
"epoch": 1.76,
"learning_rate": 2.2893954410307238e-05,
"loss": 0.0745,
"step": 660
},
{
"epoch": 1.79,
"learning_rate": 2.2398414271556e-05,
"loss": 0.0903,
"step": 670
},
{
"epoch": 1.82,
"learning_rate": 2.1902874132804756e-05,
"loss": 0.0488,
"step": 680
},
{
"epoch": 1.84,
"learning_rate": 2.1407333994053518e-05,
"loss": 0.0493,
"step": 690
},
{
"epoch": 1.87,
"learning_rate": 2.0911793855302282e-05,
"loss": 0.0452,
"step": 700
},
{
"epoch": 1.9,
"learning_rate": 2.0416253716551043e-05,
"loss": 0.0108,
"step": 710
},
{
"epoch": 1.93,
"learning_rate": 1.99207135777998e-05,
"loss": 0.0375,
"step": 720
},
{
"epoch": 1.95,
"learning_rate": 1.9425173439048562e-05,
"loss": 0.0849,
"step": 730
},
{
"epoch": 1.98,
"learning_rate": 1.8929633300297326e-05,
"loss": 0.0892,
"step": 740
},
{
"epoch": 2.0,
"eval_accuracy": 0.9896390374331551,
"eval_confusion_matrix": [
[
1501,
11
],
[
20,
1460
]
],
"eval_loss": 0.031119678169488907,
"eval_precision": 0.9896560969383656,
"eval_recall": 0.9896390374331551,
"eval_runtime": 59.6538,
"eval_samples_per_second": 50.156,
"eval_steps_per_second": 6.27,
"step": 748
},
{
"epoch": 2.01,
"learning_rate": 1.8434093161546088e-05,
"loss": 0.0562,
"step": 750
},
{
"epoch": 2.03,
"learning_rate": 1.7938553022794845e-05,
"loss": 0.0565,
"step": 760
},
{
"epoch": 2.06,
"learning_rate": 1.7443012884043606e-05,
"loss": 0.0485,
"step": 770
},
{
"epoch": 2.09,
"learning_rate": 1.694747274529237e-05,
"loss": 0.0849,
"step": 780
},
{
"epoch": 2.11,
"learning_rate": 1.6451932606541132e-05,
"loss": 0.0373,
"step": 790
},
{
"epoch": 2.14,
"learning_rate": 1.595639246778989e-05,
"loss": 0.0525,
"step": 800
},
{
"epoch": 2.17,
"learning_rate": 1.5460852329038654e-05,
"loss": 0.0849,
"step": 810
},
{
"epoch": 2.19,
"learning_rate": 1.4965312190287414e-05,
"loss": 0.0498,
"step": 820
},
{
"epoch": 2.22,
"learning_rate": 1.4469772051536176e-05,
"loss": 0.1107,
"step": 830
},
{
"epoch": 2.25,
"learning_rate": 1.3974231912784936e-05,
"loss": 0.0867,
"step": 840
},
{
"epoch": 2.27,
"learning_rate": 1.3478691774033697e-05,
"loss": 0.0509,
"step": 850
},
{
"epoch": 2.3,
"learning_rate": 1.298315163528246e-05,
"loss": 0.0694,
"step": 860
},
{
"epoch": 2.33,
"learning_rate": 1.2487611496531219e-05,
"loss": 0.0874,
"step": 870
},
{
"epoch": 2.35,
"learning_rate": 1.1992071357779982e-05,
"loss": 0.0715,
"step": 880
},
{
"epoch": 2.38,
"learning_rate": 1.1496531219028741e-05,
"loss": 0.0506,
"step": 890
},
{
"epoch": 2.41,
"learning_rate": 1.1000991080277504e-05,
"loss": 0.0458,
"step": 900
},
{
"epoch": 2.43,
"learning_rate": 1.0505450941526264e-05,
"loss": 0.0395,
"step": 910
},
{
"epoch": 2.46,
"learning_rate": 1.0009910802775026e-05,
"loss": 0.0458,
"step": 920
},
{
"epoch": 2.49,
"learning_rate": 9.514370664023786e-06,
"loss": 0.0635,
"step": 930
},
{
"epoch": 2.51,
"learning_rate": 9.018830525272549e-06,
"loss": 0.0213,
"step": 940
},
{
"epoch": 2.54,
"learning_rate": 8.523290386521308e-06,
"loss": 0.0512,
"step": 950
},
{
"epoch": 2.57,
"learning_rate": 8.02775024777007e-06,
"loss": 0.0477,
"step": 960
},
{
"epoch": 2.59,
"learning_rate": 7.532210109018831e-06,
"loss": 0.0503,
"step": 970
},
{
"epoch": 2.62,
"learning_rate": 7.036669970267593e-06,
"loss": 0.0419,
"step": 980
},
{
"epoch": 2.65,
"learning_rate": 6.541129831516353e-06,
"loss": 0.0408,
"step": 990
},
{
"epoch": 2.67,
"learning_rate": 6.045589692765114e-06,
"loss": 0.0294,
"step": 1000
},
{
"epoch": 2.7,
"learning_rate": 5.5500495540138754e-06,
"loss": 0.1206,
"step": 1010
},
{
"epoch": 2.73,
"learning_rate": 5.0545094152626366e-06,
"loss": 0.0474,
"step": 1020
},
{
"epoch": 2.75,
"learning_rate": 4.558969276511398e-06,
"loss": 0.0712,
"step": 1030
},
{
"epoch": 2.78,
"learning_rate": 4.063429137760159e-06,
"loss": 0.0973,
"step": 1040
},
{
"epoch": 2.81,
"learning_rate": 3.56788899900892e-06,
"loss": 0.0461,
"step": 1050
},
{
"epoch": 2.83,
"learning_rate": 3.072348860257681e-06,
"loss": 0.0548,
"step": 1060
},
{
"epoch": 2.86,
"learning_rate": 2.576808721506442e-06,
"loss": 0.0238,
"step": 1070
},
{
"epoch": 2.89,
"learning_rate": 2.081268582755203e-06,
"loss": 0.101,
"step": 1080
},
{
"epoch": 2.91,
"learning_rate": 1.5857284440039643e-06,
"loss": 0.0419,
"step": 1090
},
{
"epoch": 2.94,
"learning_rate": 1.0901883052527254e-06,
"loss": 0.0186,
"step": 1100
},
{
"epoch": 2.97,
"learning_rate": 5.946481665014866e-07,
"loss": 0.0123,
"step": 1110
},
{
"epoch": 2.99,
"learning_rate": 9.910802775024777e-08,
"loss": 0.0071,
"step": 1120
},
{
"epoch": 3.0,
"eval_accuracy": 0.9883021390374331,
"eval_confusion_matrix": [
[
1503,
9
],
[
26,
1454
]
],
"eval_loss": 0.03631703183054924,
"eval_precision": 0.9883638028735799,
"eval_recall": 0.9883021390374331,
"eval_runtime": 59.2122,
"eval_samples_per_second": 50.53,
"eval_steps_per_second": 6.316,
"step": 1122
},
{
"epoch": 3.0,
"step": 1122,
"total_flos": 8.923548605812163e+17,
"train_loss": 0.07081796577509074,
"train_runtime": 1074.2814,
"train_samples_per_second": 33.419,
"train_steps_per_second": 1.044
}
],
"logging_steps": 10,
"max_steps": 1122,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 8.923548605812163e+17,
"trial_name": null,
"trial_params": null
}