jaycamper's picture
End of training
dab803f
{
"best_metric": 0.8154,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-cifar100/checkpoint-1053",
"epoch": 2.9936034115138592,
"global_step": 1053,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 4.716981132075472e-06,
"loss": 4.671,
"step": 10
},
{
"epoch": 0.06,
"learning_rate": 9.433962264150944e-06,
"loss": 4.6369,
"step": 20
},
{
"epoch": 0.09,
"learning_rate": 1.4150943396226415e-05,
"loss": 4.61,
"step": 30
},
{
"epoch": 0.11,
"learning_rate": 1.8867924528301888e-05,
"loss": 4.5834,
"step": 40
},
{
"epoch": 0.14,
"learning_rate": 2.358490566037736e-05,
"loss": 4.5559,
"step": 50
},
{
"epoch": 0.17,
"learning_rate": 2.830188679245283e-05,
"loss": 4.5023,
"step": 60
},
{
"epoch": 0.2,
"learning_rate": 3.30188679245283e-05,
"loss": 4.4258,
"step": 70
},
{
"epoch": 0.23,
"learning_rate": 3.7735849056603776e-05,
"loss": 4.3076,
"step": 80
},
{
"epoch": 0.26,
"learning_rate": 4.245283018867925e-05,
"loss": 4.1446,
"step": 90
},
{
"epoch": 0.28,
"learning_rate": 4.716981132075472e-05,
"loss": 3.889,
"step": 100
},
{
"epoch": 0.31,
"learning_rate": 4.978880675818374e-05,
"loss": 3.548,
"step": 110
},
{
"epoch": 0.34,
"learning_rate": 4.9260823653643085e-05,
"loss": 3.2286,
"step": 120
},
{
"epoch": 0.37,
"learning_rate": 4.8732840549102435e-05,
"loss": 2.9794,
"step": 130
},
{
"epoch": 0.4,
"learning_rate": 4.820485744456177e-05,
"loss": 2.7993,
"step": 140
},
{
"epoch": 0.43,
"learning_rate": 4.767687434002112e-05,
"loss": 2.6407,
"step": 150
},
{
"epoch": 0.45,
"learning_rate": 4.7148891235480466e-05,
"loss": 2.4699,
"step": 160
},
{
"epoch": 0.48,
"learning_rate": 4.662090813093981e-05,
"loss": 2.4092,
"step": 170
},
{
"epoch": 0.51,
"learning_rate": 4.609292502639916e-05,
"loss": 2.2196,
"step": 180
},
{
"epoch": 0.54,
"learning_rate": 4.55649419218585e-05,
"loss": 2.2073,
"step": 190
},
{
"epoch": 0.57,
"learning_rate": 4.503695881731785e-05,
"loss": 2.1195,
"step": 200
},
{
"epoch": 0.6,
"learning_rate": 4.45089757127772e-05,
"loss": 2.0858,
"step": 210
},
{
"epoch": 0.63,
"learning_rate": 4.398099260823654e-05,
"loss": 2.0394,
"step": 220
},
{
"epoch": 0.65,
"learning_rate": 4.3453009503695884e-05,
"loss": 1.9503,
"step": 230
},
{
"epoch": 0.68,
"learning_rate": 4.292502639915523e-05,
"loss": 1.8136,
"step": 240
},
{
"epoch": 0.71,
"learning_rate": 4.239704329461457e-05,
"loss": 1.8231,
"step": 250
},
{
"epoch": 0.74,
"learning_rate": 4.186906019007392e-05,
"loss": 1.8702,
"step": 260
},
{
"epoch": 0.77,
"learning_rate": 4.1341077085533265e-05,
"loss": 1.7168,
"step": 270
},
{
"epoch": 0.8,
"learning_rate": 4.081309398099261e-05,
"loss": 1.7159,
"step": 280
},
{
"epoch": 0.82,
"learning_rate": 4.028511087645195e-05,
"loss": 1.7149,
"step": 290
},
{
"epoch": 0.85,
"learning_rate": 3.97571277719113e-05,
"loss": 1.7352,
"step": 300
},
{
"epoch": 0.88,
"learning_rate": 3.9229144667370646e-05,
"loss": 1.6819,
"step": 310
},
{
"epoch": 0.91,
"learning_rate": 3.870116156282999e-05,
"loss": 1.737,
"step": 320
},
{
"epoch": 0.94,
"learning_rate": 3.817317845828934e-05,
"loss": 1.6592,
"step": 330
},
{
"epoch": 0.97,
"learning_rate": 3.764519535374868e-05,
"loss": 1.6503,
"step": 340
},
{
"epoch": 1.0,
"learning_rate": 3.711721224920803e-05,
"loss": 1.5918,
"step": 350
},
{
"epoch": 1.0,
"eval_accuracy": 0.7364,
"eval_loss": 0.9103603363037109,
"eval_runtime": 29.7892,
"eval_samples_per_second": 167.846,
"eval_steps_per_second": 5.27,
"step": 351
},
{
"epoch": 1.02,
"learning_rate": 3.658922914466738e-05,
"loss": 1.5672,
"step": 360
},
{
"epoch": 1.05,
"learning_rate": 3.6061246040126714e-05,
"loss": 1.6706,
"step": 370
},
{
"epoch": 1.08,
"learning_rate": 3.5533262935586064e-05,
"loss": 1.5448,
"step": 380
},
{
"epoch": 1.11,
"learning_rate": 3.500527983104541e-05,
"loss": 1.5142,
"step": 390
},
{
"epoch": 1.14,
"learning_rate": 3.447729672650475e-05,
"loss": 1.4882,
"step": 400
},
{
"epoch": 1.17,
"learning_rate": 3.3949313621964095e-05,
"loss": 1.4584,
"step": 410
},
{
"epoch": 1.19,
"learning_rate": 3.3421330517423445e-05,
"loss": 1.391,
"step": 420
},
{
"epoch": 1.22,
"learning_rate": 3.289334741288279e-05,
"loss": 1.4512,
"step": 430
},
{
"epoch": 1.25,
"learning_rate": 3.236536430834213e-05,
"loss": 1.535,
"step": 440
},
{
"epoch": 1.28,
"learning_rate": 3.183738120380148e-05,
"loss": 1.4058,
"step": 450
},
{
"epoch": 1.31,
"learning_rate": 3.130939809926082e-05,
"loss": 1.4024,
"step": 460
},
{
"epoch": 1.34,
"learning_rate": 3.078141499472017e-05,
"loss": 1.3875,
"step": 470
},
{
"epoch": 1.36,
"learning_rate": 3.0253431890179517e-05,
"loss": 1.4125,
"step": 480
},
{
"epoch": 1.39,
"learning_rate": 2.972544878563886e-05,
"loss": 1.4099,
"step": 490
},
{
"epoch": 1.42,
"learning_rate": 2.9197465681098207e-05,
"loss": 1.3635,
"step": 500
},
{
"epoch": 1.45,
"learning_rate": 2.8669482576557548e-05,
"loss": 1.3427,
"step": 510
},
{
"epoch": 1.48,
"learning_rate": 2.8141499472016898e-05,
"loss": 1.3257,
"step": 520
},
{
"epoch": 1.51,
"learning_rate": 2.7613516367476245e-05,
"loss": 1.3757,
"step": 530
},
{
"epoch": 1.54,
"learning_rate": 2.7085533262935585e-05,
"loss": 1.3744,
"step": 540
},
{
"epoch": 1.56,
"learning_rate": 2.6557550158394935e-05,
"loss": 1.4361,
"step": 550
},
{
"epoch": 1.59,
"learning_rate": 2.6029567053854276e-05,
"loss": 1.3881,
"step": 560
},
{
"epoch": 1.62,
"learning_rate": 2.5501583949313622e-05,
"loss": 1.2947,
"step": 570
},
{
"epoch": 1.65,
"learning_rate": 2.497360084477297e-05,
"loss": 1.3397,
"step": 580
},
{
"epoch": 1.68,
"learning_rate": 2.4445617740232313e-05,
"loss": 1.3136,
"step": 590
},
{
"epoch": 1.71,
"learning_rate": 2.391763463569166e-05,
"loss": 1.3448,
"step": 600
},
{
"epoch": 1.73,
"learning_rate": 2.3389651531151003e-05,
"loss": 1.2951,
"step": 610
},
{
"epoch": 1.76,
"learning_rate": 2.286166842661035e-05,
"loss": 1.3231,
"step": 620
},
{
"epoch": 1.79,
"learning_rate": 2.2333685322069694e-05,
"loss": 1.3093,
"step": 630
},
{
"epoch": 1.82,
"learning_rate": 2.180570221752904e-05,
"loss": 1.3218,
"step": 640
},
{
"epoch": 1.85,
"learning_rate": 2.1277719112988384e-05,
"loss": 1.3456,
"step": 650
},
{
"epoch": 1.88,
"learning_rate": 2.074973600844773e-05,
"loss": 1.3029,
"step": 660
},
{
"epoch": 1.9,
"learning_rate": 2.0221752903907075e-05,
"loss": 1.2968,
"step": 670
},
{
"epoch": 1.93,
"learning_rate": 1.9693769799366422e-05,
"loss": 1.2704,
"step": 680
},
{
"epoch": 1.96,
"learning_rate": 1.9165786694825765e-05,
"loss": 1.2607,
"step": 690
},
{
"epoch": 1.99,
"learning_rate": 1.863780359028511e-05,
"loss": 1.2837,
"step": 700
},
{
"epoch": 2.0,
"eval_accuracy": 0.796,
"eval_loss": 0.6690887808799744,
"eval_runtime": 29.9082,
"eval_samples_per_second": 167.178,
"eval_steps_per_second": 5.249,
"step": 703
},
{
"epoch": 2.02,
"learning_rate": 1.810982048574446e-05,
"loss": 1.3087,
"step": 710
},
{
"epoch": 2.05,
"learning_rate": 1.7581837381203803e-05,
"loss": 1.2362,
"step": 720
},
{
"epoch": 2.08,
"learning_rate": 1.7053854276663146e-05,
"loss": 1.1623,
"step": 730
},
{
"epoch": 2.1,
"learning_rate": 1.6525871172122493e-05,
"loss": 1.2917,
"step": 740
},
{
"epoch": 2.13,
"learning_rate": 1.5997888067581837e-05,
"loss": 1.1616,
"step": 750
},
{
"epoch": 2.16,
"learning_rate": 1.5469904963041184e-05,
"loss": 1.1987,
"step": 760
},
{
"epoch": 2.19,
"learning_rate": 1.4941921858500529e-05,
"loss": 1.1655,
"step": 770
},
{
"epoch": 2.22,
"learning_rate": 1.4413938753959874e-05,
"loss": 1.1648,
"step": 780
},
{
"epoch": 2.25,
"learning_rate": 1.388595564941922e-05,
"loss": 1.2011,
"step": 790
},
{
"epoch": 2.27,
"learning_rate": 1.3357972544878563e-05,
"loss": 1.1683,
"step": 800
},
{
"epoch": 2.3,
"learning_rate": 1.2829989440337912e-05,
"loss": 1.2133,
"step": 810
},
{
"epoch": 2.33,
"learning_rate": 1.2302006335797255e-05,
"loss": 1.1571,
"step": 820
},
{
"epoch": 2.36,
"learning_rate": 1.17740232312566e-05,
"loss": 1.3042,
"step": 830
},
{
"epoch": 2.39,
"learning_rate": 1.1246040126715946e-05,
"loss": 1.1284,
"step": 840
},
{
"epoch": 2.42,
"learning_rate": 1.0718057022175291e-05,
"loss": 1.1856,
"step": 850
},
{
"epoch": 2.44,
"learning_rate": 1.0190073917634636e-05,
"loss": 1.2035,
"step": 860
},
{
"epoch": 2.47,
"learning_rate": 9.662090813093982e-06,
"loss": 1.1771,
"step": 870
},
{
"epoch": 2.5,
"learning_rate": 9.134107708553327e-06,
"loss": 1.1834,
"step": 880
},
{
"epoch": 2.53,
"learning_rate": 8.606124604012672e-06,
"loss": 1.1581,
"step": 890
},
{
"epoch": 2.56,
"learning_rate": 8.078141499472017e-06,
"loss": 1.2464,
"step": 900
},
{
"epoch": 2.59,
"learning_rate": 7.5501583949313625e-06,
"loss": 1.2024,
"step": 910
},
{
"epoch": 2.62,
"learning_rate": 7.022175290390708e-06,
"loss": 1.1645,
"step": 920
},
{
"epoch": 2.64,
"learning_rate": 6.494192185850054e-06,
"loss": 1.1576,
"step": 930
},
{
"epoch": 2.67,
"learning_rate": 5.966209081309398e-06,
"loss": 1.1729,
"step": 940
},
{
"epoch": 2.7,
"learning_rate": 5.438225976768744e-06,
"loss": 1.1783,
"step": 950
},
{
"epoch": 2.73,
"learning_rate": 4.910242872228089e-06,
"loss": 1.1083,
"step": 960
},
{
"epoch": 2.76,
"learning_rate": 4.382259767687434e-06,
"loss": 1.1666,
"step": 970
},
{
"epoch": 2.79,
"learning_rate": 3.854276663146779e-06,
"loss": 1.1307,
"step": 980
},
{
"epoch": 2.81,
"learning_rate": 3.326293558606125e-06,
"loss": 1.1937,
"step": 990
},
{
"epoch": 2.84,
"learning_rate": 2.79831045406547e-06,
"loss": 1.1359,
"step": 1000
},
{
"epoch": 2.87,
"learning_rate": 2.2703273495248154e-06,
"loss": 1.0982,
"step": 1010
},
{
"epoch": 2.9,
"learning_rate": 1.7423442449841606e-06,
"loss": 1.1268,
"step": 1020
},
{
"epoch": 2.93,
"learning_rate": 1.2143611404435059e-06,
"loss": 1.1451,
"step": 1030
},
{
"epoch": 2.96,
"learning_rate": 6.863780359028511e-07,
"loss": 1.1403,
"step": 1040
},
{
"epoch": 2.99,
"learning_rate": 1.5839493136219642e-07,
"loss": 1.1241,
"step": 1050
},
{
"epoch": 2.99,
"eval_accuracy": 0.8154,
"eval_loss": 0.5995984673500061,
"eval_runtime": 29.597,
"eval_samples_per_second": 168.936,
"eval_steps_per_second": 5.305,
"step": 1053
},
{
"epoch": 2.99,
"step": 1053,
"total_flos": 3.3581672767780946e+18,
"train_loss": 1.784397272416103,
"train_runtime": 2010.7817,
"train_samples_per_second": 67.138,
"train_steps_per_second": 0.524
}
],
"max_steps": 1053,
"num_train_epochs": 3,
"total_flos": 3.3581672767780946e+18,
"trial_name": null,
"trial_params": null
}