julianz1's picture
End of training
c02540f verified
{
"best_metric": 0.7760274575376221,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-food101/checkpoint-1509",
"epoch": 2.9985096870342773,
"eval_steps": 500,
"global_step": 1509,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 3.3112582781456956e-06,
"loss": 4.6663,
"step": 10
},
{
"epoch": 0.04,
"learning_rate": 6.622516556291391e-06,
"loss": 4.6531,
"step": 20
},
{
"epoch": 0.06,
"learning_rate": 9.933774834437086e-06,
"loss": 4.628,
"step": 30
},
{
"epoch": 0.08,
"learning_rate": 1.3245033112582782e-05,
"loss": 4.613,
"step": 40
},
{
"epoch": 0.1,
"learning_rate": 1.655629139072848e-05,
"loss": 4.5875,
"step": 50
},
{
"epoch": 0.12,
"learning_rate": 1.9867549668874173e-05,
"loss": 4.5399,
"step": 60
},
{
"epoch": 0.14,
"learning_rate": 2.317880794701987e-05,
"loss": 4.5013,
"step": 70
},
{
"epoch": 0.16,
"learning_rate": 2.6490066225165565e-05,
"loss": 4.42,
"step": 80
},
{
"epoch": 0.18,
"learning_rate": 2.980132450331126e-05,
"loss": 4.298,
"step": 90
},
{
"epoch": 0.2,
"learning_rate": 3.311258278145696e-05,
"loss": 4.1074,
"step": 100
},
{
"epoch": 0.22,
"learning_rate": 3.6423841059602646e-05,
"loss": 3.9249,
"step": 110
},
{
"epoch": 0.24,
"learning_rate": 3.9735099337748346e-05,
"loss": 3.6311,
"step": 120
},
{
"epoch": 0.26,
"learning_rate": 4.304635761589404e-05,
"loss": 3.4282,
"step": 130
},
{
"epoch": 0.28,
"learning_rate": 4.635761589403974e-05,
"loss": 3.1668,
"step": 140
},
{
"epoch": 0.3,
"learning_rate": 4.966887417218543e-05,
"loss": 3.0268,
"step": 150
},
{
"epoch": 0.32,
"learning_rate": 4.966863033873343e-05,
"loss": 2.7985,
"step": 160
},
{
"epoch": 0.34,
"learning_rate": 4.9300441826215024e-05,
"loss": 2.6903,
"step": 170
},
{
"epoch": 0.36,
"learning_rate": 4.8932253313696616e-05,
"loss": 2.4989,
"step": 180
},
{
"epoch": 0.38,
"learning_rate": 4.856406480117821e-05,
"loss": 2.5033,
"step": 190
},
{
"epoch": 0.4,
"learning_rate": 4.8195876288659794e-05,
"loss": 2.3169,
"step": 200
},
{
"epoch": 0.42,
"learning_rate": 4.7827687776141386e-05,
"loss": 2.3053,
"step": 210
},
{
"epoch": 0.44,
"learning_rate": 4.745949926362298e-05,
"loss": 2.1974,
"step": 220
},
{
"epoch": 0.46,
"learning_rate": 4.709131075110457e-05,
"loss": 2.0775,
"step": 230
},
{
"epoch": 0.48,
"learning_rate": 4.6723122238586156e-05,
"loss": 2.0065,
"step": 240
},
{
"epoch": 0.5,
"learning_rate": 4.635493372606775e-05,
"loss": 2.0521,
"step": 250
},
{
"epoch": 0.52,
"learning_rate": 4.598674521354934e-05,
"loss": 1.9464,
"step": 260
},
{
"epoch": 0.54,
"learning_rate": 4.561855670103093e-05,
"loss": 2.0498,
"step": 270
},
{
"epoch": 0.56,
"learning_rate": 4.525036818851252e-05,
"loss": 1.9544,
"step": 280
},
{
"epoch": 0.58,
"learning_rate": 4.488217967599411e-05,
"loss": 1.902,
"step": 290
},
{
"epoch": 0.6,
"learning_rate": 4.45139911634757e-05,
"loss": 1.8034,
"step": 300
},
{
"epoch": 0.62,
"learning_rate": 4.4145802650957294e-05,
"loss": 1.7483,
"step": 310
},
{
"epoch": 0.64,
"learning_rate": 4.3777614138438886e-05,
"loss": 1.8315,
"step": 320
},
{
"epoch": 0.66,
"learning_rate": 4.340942562592047e-05,
"loss": 1.7785,
"step": 330
},
{
"epoch": 0.68,
"learning_rate": 4.3041237113402064e-05,
"loss": 1.85,
"step": 340
},
{
"epoch": 0.7,
"learning_rate": 4.2673048600883656e-05,
"loss": 1.7037,
"step": 350
},
{
"epoch": 0.72,
"learning_rate": 4.230486008836525e-05,
"loss": 1.6452,
"step": 360
},
{
"epoch": 0.74,
"learning_rate": 4.193667157584683e-05,
"loss": 1.6515,
"step": 370
},
{
"epoch": 0.76,
"learning_rate": 4.1568483063328425e-05,
"loss": 1.6375,
"step": 380
},
{
"epoch": 0.77,
"learning_rate": 4.120029455081002e-05,
"loss": 1.6522,
"step": 390
},
{
"epoch": 0.79,
"learning_rate": 4.083210603829161e-05,
"loss": 1.5982,
"step": 400
},
{
"epoch": 0.81,
"learning_rate": 4.0463917525773195e-05,
"loss": 1.5984,
"step": 410
},
{
"epoch": 0.83,
"learning_rate": 4.009572901325479e-05,
"loss": 1.5961,
"step": 420
},
{
"epoch": 0.85,
"learning_rate": 3.972754050073638e-05,
"loss": 1.6263,
"step": 430
},
{
"epoch": 0.87,
"learning_rate": 3.935935198821797e-05,
"loss": 1.5302,
"step": 440
},
{
"epoch": 0.89,
"learning_rate": 3.899116347569956e-05,
"loss": 1.5545,
"step": 450
},
{
"epoch": 0.91,
"learning_rate": 3.862297496318115e-05,
"loss": 1.5918,
"step": 460
},
{
"epoch": 0.93,
"learning_rate": 3.825478645066274e-05,
"loss": 1.539,
"step": 470
},
{
"epoch": 0.95,
"learning_rate": 3.7886597938144333e-05,
"loss": 1.5363,
"step": 480
},
{
"epoch": 0.97,
"learning_rate": 3.7518409425625926e-05,
"loss": 1.5359,
"step": 490
},
{
"epoch": 0.99,
"learning_rate": 3.715022091310751e-05,
"loss": 1.4777,
"step": 500
},
{
"epoch": 1.0,
"eval_accuracy": 0.7062395494147672,
"eval_loss": 1.1435766220092773,
"eval_runtime": 107.4159,
"eval_samples_per_second": 105.785,
"eval_steps_per_second": 3.314,
"step": 503
},
{
"epoch": 1.01,
"learning_rate": 3.67820324005891e-05,
"loss": 1.492,
"step": 510
},
{
"epoch": 1.03,
"learning_rate": 3.6413843888070695e-05,
"loss": 1.4226,
"step": 520
},
{
"epoch": 1.05,
"learning_rate": 3.604565537555229e-05,
"loss": 1.3538,
"step": 530
},
{
"epoch": 1.07,
"learning_rate": 3.567746686303387e-05,
"loss": 1.3888,
"step": 540
},
{
"epoch": 1.09,
"learning_rate": 3.5309278350515465e-05,
"loss": 1.3147,
"step": 550
},
{
"epoch": 1.11,
"learning_rate": 3.494108983799706e-05,
"loss": 1.414,
"step": 560
},
{
"epoch": 1.13,
"learning_rate": 3.457290132547865e-05,
"loss": 1.2426,
"step": 570
},
{
"epoch": 1.15,
"learning_rate": 3.4204712812960235e-05,
"loss": 1.3277,
"step": 580
},
{
"epoch": 1.17,
"learning_rate": 3.383652430044183e-05,
"loss": 1.3334,
"step": 590
},
{
"epoch": 1.19,
"learning_rate": 3.346833578792342e-05,
"loss": 1.3156,
"step": 600
},
{
"epoch": 1.21,
"learning_rate": 3.310014727540501e-05,
"loss": 1.2929,
"step": 610
},
{
"epoch": 1.23,
"learning_rate": 3.2731958762886596e-05,
"loss": 1.274,
"step": 620
},
{
"epoch": 1.25,
"learning_rate": 3.236377025036819e-05,
"loss": 1.3538,
"step": 630
},
{
"epoch": 1.27,
"learning_rate": 3.199558173784978e-05,
"loss": 1.3097,
"step": 640
},
{
"epoch": 1.29,
"learning_rate": 3.162739322533137e-05,
"loss": 1.3836,
"step": 650
},
{
"epoch": 1.31,
"learning_rate": 3.1259204712812965e-05,
"loss": 1.2941,
"step": 660
},
{
"epoch": 1.33,
"learning_rate": 3.089101620029455e-05,
"loss": 1.3041,
"step": 670
},
{
"epoch": 1.35,
"learning_rate": 3.052282768777614e-05,
"loss": 1.35,
"step": 680
},
{
"epoch": 1.37,
"learning_rate": 3.015463917525773e-05,
"loss": 1.2847,
"step": 690
},
{
"epoch": 1.39,
"learning_rate": 2.9786450662739324e-05,
"loss": 1.3606,
"step": 700
},
{
"epoch": 1.41,
"learning_rate": 2.9418262150220916e-05,
"loss": 1.2239,
"step": 710
},
{
"epoch": 1.43,
"learning_rate": 2.9050073637702508e-05,
"loss": 1.2794,
"step": 720
},
{
"epoch": 1.45,
"learning_rate": 2.8681885125184093e-05,
"loss": 1.2988,
"step": 730
},
{
"epoch": 1.47,
"learning_rate": 2.8313696612665685e-05,
"loss": 1.2714,
"step": 740
},
{
"epoch": 1.49,
"learning_rate": 2.7945508100147278e-05,
"loss": 1.2118,
"step": 750
},
{
"epoch": 1.51,
"learning_rate": 2.757731958762887e-05,
"loss": 1.2792,
"step": 760
},
{
"epoch": 1.53,
"learning_rate": 2.7209131075110455e-05,
"loss": 1.3307,
"step": 770
},
{
"epoch": 1.55,
"learning_rate": 2.6840942562592047e-05,
"loss": 1.2683,
"step": 780
},
{
"epoch": 1.57,
"learning_rate": 2.647275405007364e-05,
"loss": 1.2789,
"step": 790
},
{
"epoch": 1.59,
"learning_rate": 2.610456553755523e-05,
"loss": 1.261,
"step": 800
},
{
"epoch": 1.61,
"learning_rate": 2.5736377025036817e-05,
"loss": 1.3408,
"step": 810
},
{
"epoch": 1.63,
"learning_rate": 2.536818851251841e-05,
"loss": 1.27,
"step": 820
},
{
"epoch": 1.65,
"learning_rate": 2.5e-05,
"loss": 1.1915,
"step": 830
},
{
"epoch": 1.67,
"learning_rate": 2.4631811487481593e-05,
"loss": 1.1396,
"step": 840
},
{
"epoch": 1.69,
"learning_rate": 2.4263622974963182e-05,
"loss": 1.1942,
"step": 850
},
{
"epoch": 1.71,
"learning_rate": 2.3895434462444774e-05,
"loss": 1.1793,
"step": 860
},
{
"epoch": 1.73,
"learning_rate": 2.3527245949926363e-05,
"loss": 1.192,
"step": 870
},
{
"epoch": 1.75,
"learning_rate": 2.3159057437407955e-05,
"loss": 1.2951,
"step": 880
},
{
"epoch": 1.77,
"learning_rate": 2.2790868924889544e-05,
"loss": 1.2129,
"step": 890
},
{
"epoch": 1.79,
"learning_rate": 2.2422680412371136e-05,
"loss": 1.192,
"step": 900
},
{
"epoch": 1.81,
"learning_rate": 2.2054491899852725e-05,
"loss": 1.2012,
"step": 910
},
{
"epoch": 1.83,
"learning_rate": 2.1686303387334317e-05,
"loss": 1.2543,
"step": 920
},
{
"epoch": 1.85,
"learning_rate": 2.1318114874815906e-05,
"loss": 1.2318,
"step": 930
},
{
"epoch": 1.87,
"learning_rate": 2.0949926362297498e-05,
"loss": 1.223,
"step": 940
},
{
"epoch": 1.89,
"learning_rate": 2.0581737849779087e-05,
"loss": 1.2347,
"step": 950
},
{
"epoch": 1.91,
"learning_rate": 2.021354933726068e-05,
"loss": 1.2412,
"step": 960
},
{
"epoch": 1.93,
"learning_rate": 1.9845360824742268e-05,
"loss": 1.1686,
"step": 970
},
{
"epoch": 1.95,
"learning_rate": 1.947717231222386e-05,
"loss": 1.1424,
"step": 980
},
{
"epoch": 1.97,
"learning_rate": 1.910898379970545e-05,
"loss": 1.2094,
"step": 990
},
{
"epoch": 1.99,
"learning_rate": 1.874079528718704e-05,
"loss": 1.2418,
"step": 1000
},
{
"epoch": 2.0,
"eval_accuracy": 0.7571063979582857,
"eval_loss": 0.9226512908935547,
"eval_runtime": 106.3435,
"eval_samples_per_second": 106.852,
"eval_steps_per_second": 3.348,
"step": 1006
},
{
"epoch": 2.01,
"learning_rate": 1.8372606774668633e-05,
"loss": 1.2256,
"step": 1010
},
{
"epoch": 2.03,
"learning_rate": 1.800441826215022e-05,
"loss": 1.1317,
"step": 1020
},
{
"epoch": 2.05,
"learning_rate": 1.7636229749631814e-05,
"loss": 1.1062,
"step": 1030
},
{
"epoch": 2.07,
"learning_rate": 1.7268041237113403e-05,
"loss": 1.1442,
"step": 1040
},
{
"epoch": 2.09,
"learning_rate": 1.6899852724594995e-05,
"loss": 1.0836,
"step": 1050
},
{
"epoch": 2.11,
"learning_rate": 1.6531664212076583e-05,
"loss": 1.1139,
"step": 1060
},
{
"epoch": 2.13,
"learning_rate": 1.6163475699558176e-05,
"loss": 1.0567,
"step": 1070
},
{
"epoch": 2.15,
"learning_rate": 1.5795287187039764e-05,
"loss": 1.0716,
"step": 1080
},
{
"epoch": 2.17,
"learning_rate": 1.5427098674521357e-05,
"loss": 1.1631,
"step": 1090
},
{
"epoch": 2.19,
"learning_rate": 1.5058910162002945e-05,
"loss": 1.0727,
"step": 1100
},
{
"epoch": 2.21,
"learning_rate": 1.4690721649484537e-05,
"loss": 1.1132,
"step": 1110
},
{
"epoch": 2.23,
"learning_rate": 1.4322533136966126e-05,
"loss": 1.0497,
"step": 1120
},
{
"epoch": 2.25,
"learning_rate": 1.3954344624447718e-05,
"loss": 1.1202,
"step": 1130
},
{
"epoch": 2.27,
"learning_rate": 1.3586156111929307e-05,
"loss": 1.0817,
"step": 1140
},
{
"epoch": 2.29,
"learning_rate": 1.32179675994109e-05,
"loss": 1.1048,
"step": 1150
},
{
"epoch": 2.31,
"learning_rate": 1.2849779086892488e-05,
"loss": 1.1784,
"step": 1160
},
{
"epoch": 2.32,
"learning_rate": 1.248159057437408e-05,
"loss": 1.08,
"step": 1170
},
{
"epoch": 2.34,
"learning_rate": 1.211340206185567e-05,
"loss": 1.1461,
"step": 1180
},
{
"epoch": 2.36,
"learning_rate": 1.1745213549337261e-05,
"loss": 1.0777,
"step": 1190
},
{
"epoch": 2.38,
"learning_rate": 1.1377025036818852e-05,
"loss": 1.0184,
"step": 1200
},
{
"epoch": 2.4,
"learning_rate": 1.1008836524300442e-05,
"loss": 1.0455,
"step": 1210
},
{
"epoch": 2.42,
"learning_rate": 1.0640648011782033e-05,
"loss": 1.095,
"step": 1220
},
{
"epoch": 2.44,
"learning_rate": 1.0272459499263623e-05,
"loss": 1.1394,
"step": 1230
},
{
"epoch": 2.46,
"learning_rate": 9.904270986745213e-06,
"loss": 1.1076,
"step": 1240
},
{
"epoch": 2.48,
"learning_rate": 9.536082474226804e-06,
"loss": 1.0063,
"step": 1250
},
{
"epoch": 2.5,
"learning_rate": 9.167893961708394e-06,
"loss": 1.1044,
"step": 1260
},
{
"epoch": 2.52,
"learning_rate": 8.799705449189986e-06,
"loss": 1.041,
"step": 1270
},
{
"epoch": 2.54,
"learning_rate": 8.431516936671577e-06,
"loss": 1.0551,
"step": 1280
},
{
"epoch": 2.56,
"learning_rate": 8.063328424153167e-06,
"loss": 1.0601,
"step": 1290
},
{
"epoch": 2.58,
"learning_rate": 7.695139911634758e-06,
"loss": 1.1787,
"step": 1300
},
{
"epoch": 2.6,
"learning_rate": 7.326951399116348e-06,
"loss": 1.0445,
"step": 1310
},
{
"epoch": 2.62,
"learning_rate": 6.958762886597939e-06,
"loss": 1.105,
"step": 1320
},
{
"epoch": 2.64,
"learning_rate": 6.590574374079529e-06,
"loss": 1.0418,
"step": 1330
},
{
"epoch": 2.66,
"learning_rate": 6.22238586156112e-06,
"loss": 0.9967,
"step": 1340
},
{
"epoch": 2.68,
"learning_rate": 5.85419734904271e-06,
"loss": 1.0506,
"step": 1350
},
{
"epoch": 2.7,
"learning_rate": 5.486008836524301e-06,
"loss": 1.1154,
"step": 1360
},
{
"epoch": 2.72,
"learning_rate": 5.117820324005891e-06,
"loss": 1.058,
"step": 1370
},
{
"epoch": 2.74,
"learning_rate": 4.7496318114874815e-06,
"loss": 1.0997,
"step": 1380
},
{
"epoch": 2.76,
"learning_rate": 4.381443298969072e-06,
"loss": 1.0751,
"step": 1390
},
{
"epoch": 2.78,
"learning_rate": 4.013254786450663e-06,
"loss": 1.0658,
"step": 1400
},
{
"epoch": 2.8,
"learning_rate": 3.6450662739322538e-06,
"loss": 1.0243,
"step": 1410
},
{
"epoch": 2.82,
"learning_rate": 3.2768777614138442e-06,
"loss": 1.003,
"step": 1420
},
{
"epoch": 2.84,
"learning_rate": 2.9086892488954347e-06,
"loss": 1.0251,
"step": 1430
},
{
"epoch": 2.86,
"learning_rate": 2.540500736377025e-06,
"loss": 1.0049,
"step": 1440
},
{
"epoch": 2.88,
"learning_rate": 2.1723122238586156e-06,
"loss": 1.1317,
"step": 1450
},
{
"epoch": 2.9,
"learning_rate": 1.804123711340206e-06,
"loss": 1.063,
"step": 1460
},
{
"epoch": 2.92,
"learning_rate": 1.4359351988217967e-06,
"loss": 1.0558,
"step": 1470
},
{
"epoch": 2.94,
"learning_rate": 1.0677466863033874e-06,
"loss": 1.0737,
"step": 1480
},
{
"epoch": 2.96,
"learning_rate": 6.99558173784978e-07,
"loss": 1.0705,
"step": 1490
},
{
"epoch": 2.98,
"learning_rate": 3.313696612665685e-07,
"loss": 1.0657,
"step": 1500
},
{
"epoch": 3.0,
"eval_accuracy": 0.7760274575376221,
"eval_loss": 0.8486079573631287,
"eval_runtime": 106.6601,
"eval_samples_per_second": 106.535,
"eval_steps_per_second": 3.338,
"step": 1509
},
{
"epoch": 3.0,
"step": 1509,
"total_flos": 4.812819303609225e+18,
"train_loss": 1.6366877293570932,
"train_runtime": 3640.8494,
"train_samples_per_second": 53.054,
"train_steps_per_second": 0.414
},
{
"epoch": 3.0,
"eval_accuracy": 0.7760274575376221,
"eval_loss": 0.8486079573631287,
"eval_runtime": 130.1364,
"eval_samples_per_second": 87.316,
"eval_steps_per_second": 2.736,
"step": 1509
}
],
"logging_steps": 10,
"max_steps": 1509,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 4.812819303609225e+18,
"trial_name": null,
"trial_params": null
}