vit_receipts_classifier / trainer_state.json
jjmcarrascosa's picture
add model
6a3f653
{
"best_metric": 0.011647290550172329,
"best_model_checkpoint": "./vit_tickers_binaryclf/checkpoint-1000",
"epoch": 1.0,
"global_step": 1785,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 0.0001988795518207283,
"loss": 0.1948,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 0.0001977591036414566,
"loss": 0.0625,
"step": 20
},
{
"epoch": 0.02,
"learning_rate": 0.00019663865546218486,
"loss": 0.018,
"step": 30
},
{
"epoch": 0.02,
"learning_rate": 0.00019551820728291316,
"loss": 0.0079,
"step": 40
},
{
"epoch": 0.03,
"learning_rate": 0.00019439775910364145,
"loss": 0.0311,
"step": 50
},
{
"epoch": 0.03,
"learning_rate": 0.00019327731092436975,
"loss": 0.0072,
"step": 60
},
{
"epoch": 0.04,
"learning_rate": 0.00019215686274509807,
"loss": 0.0053,
"step": 70
},
{
"epoch": 0.04,
"learning_rate": 0.00019103641456582636,
"loss": 0.0037,
"step": 80
},
{
"epoch": 0.05,
"learning_rate": 0.00018991596638655463,
"loss": 0.0038,
"step": 90
},
{
"epoch": 0.06,
"learning_rate": 0.00018879551820728292,
"loss": 0.004,
"step": 100
},
{
"epoch": 0.06,
"learning_rate": 0.00018767507002801122,
"loss": 0.244,
"step": 110
},
{
"epoch": 0.07,
"learning_rate": 0.0001865546218487395,
"loss": 0.1787,
"step": 120
},
{
"epoch": 0.07,
"learning_rate": 0.0001854341736694678,
"loss": 0.084,
"step": 130
},
{
"epoch": 0.08,
"learning_rate": 0.00018431372549019607,
"loss": 0.0118,
"step": 140
},
{
"epoch": 0.08,
"learning_rate": 0.00018319327731092437,
"loss": 0.0053,
"step": 150
},
{
"epoch": 0.09,
"learning_rate": 0.00018207282913165266,
"loss": 0.1699,
"step": 160
},
{
"epoch": 0.1,
"learning_rate": 0.00018095238095238095,
"loss": 0.1418,
"step": 170
},
{
"epoch": 0.1,
"learning_rate": 0.00017983193277310925,
"loss": 0.0859,
"step": 180
},
{
"epoch": 0.11,
"learning_rate": 0.00017871148459383754,
"loss": 0.0463,
"step": 190
},
{
"epoch": 0.11,
"learning_rate": 0.00017759103641456584,
"loss": 0.0699,
"step": 200
},
{
"epoch": 0.12,
"learning_rate": 0.00017647058823529413,
"loss": 0.0116,
"step": 210
},
{
"epoch": 0.12,
"learning_rate": 0.00017535014005602242,
"loss": 0.0046,
"step": 220
},
{
"epoch": 0.13,
"learning_rate": 0.00017422969187675072,
"loss": 0.076,
"step": 230
},
{
"epoch": 0.13,
"learning_rate": 0.000173109243697479,
"loss": 0.0159,
"step": 240
},
{
"epoch": 0.14,
"learning_rate": 0.0001719887955182073,
"loss": 0.0372,
"step": 250
},
{
"epoch": 0.15,
"learning_rate": 0.00017086834733893557,
"loss": 0.0603,
"step": 260
},
{
"epoch": 0.15,
"learning_rate": 0.00016974789915966387,
"loss": 0.0581,
"step": 270
},
{
"epoch": 0.16,
"learning_rate": 0.00016862745098039216,
"loss": 0.0122,
"step": 280
},
{
"epoch": 0.16,
"learning_rate": 0.00016750700280112045,
"loss": 0.0411,
"step": 290
},
{
"epoch": 0.17,
"learning_rate": 0.00016638655462184875,
"loss": 0.0232,
"step": 300
},
{
"epoch": 0.17,
"learning_rate": 0.00016526610644257704,
"loss": 0.0189,
"step": 310
},
{
"epoch": 0.18,
"learning_rate": 0.00016414565826330534,
"loss": 0.0293,
"step": 320
},
{
"epoch": 0.18,
"learning_rate": 0.00016302521008403363,
"loss": 0.041,
"step": 330
},
{
"epoch": 0.19,
"learning_rate": 0.00016190476190476192,
"loss": 0.1021,
"step": 340
},
{
"epoch": 0.2,
"learning_rate": 0.00016078431372549022,
"loss": 0.0216,
"step": 350
},
{
"epoch": 0.2,
"learning_rate": 0.0001596638655462185,
"loss": 0.0747,
"step": 360
},
{
"epoch": 0.21,
"learning_rate": 0.00015854341736694678,
"loss": 0.0056,
"step": 370
},
{
"epoch": 0.21,
"learning_rate": 0.00015742296918767507,
"loss": 0.0043,
"step": 380
},
{
"epoch": 0.22,
"learning_rate": 0.00015630252100840337,
"loss": 0.0045,
"step": 390
},
{
"epoch": 0.22,
"learning_rate": 0.00015518207282913166,
"loss": 0.0449,
"step": 400
},
{
"epoch": 0.23,
"learning_rate": 0.00015417366946778713,
"loss": 0.0539,
"step": 410
},
{
"epoch": 0.24,
"learning_rate": 0.0001530532212885154,
"loss": 0.0032,
"step": 420
},
{
"epoch": 0.24,
"learning_rate": 0.0001519327731092437,
"loss": 0.0402,
"step": 430
},
{
"epoch": 0.25,
"learning_rate": 0.00015081232492997199,
"loss": 0.0336,
"step": 440
},
{
"epoch": 0.25,
"learning_rate": 0.00014969187675070028,
"loss": 0.0038,
"step": 450
},
{
"epoch": 0.26,
"learning_rate": 0.00014857142857142857,
"loss": 0.0415,
"step": 460
},
{
"epoch": 0.26,
"learning_rate": 0.00014745098039215687,
"loss": 0.0037,
"step": 470
},
{
"epoch": 0.27,
"learning_rate": 0.00014633053221288516,
"loss": 0.0028,
"step": 480
},
{
"epoch": 0.27,
"learning_rate": 0.00014521008403361346,
"loss": 0.04,
"step": 490
},
{
"epoch": 0.28,
"learning_rate": 0.00014408963585434175,
"loss": 0.0026,
"step": 500
},
{
"epoch": 0.28,
"eval_f1": 0.9981801637852593,
"eval_loss": 0.018740175291895866,
"eval_runtime": 83.1765,
"eval_samples_per_second": 67.928,
"eval_steps_per_second": 8.5,
"step": 500
},
{
"epoch": 0.29,
"learning_rate": 0.00014296918767507004,
"loss": 0.0028,
"step": 510
},
{
"epoch": 0.29,
"learning_rate": 0.00014184873949579834,
"loss": 0.0024,
"step": 520
},
{
"epoch": 0.3,
"learning_rate": 0.00014072829131652663,
"loss": 0.0023,
"step": 530
},
{
"epoch": 0.3,
"learning_rate": 0.0001396078431372549,
"loss": 0.0021,
"step": 540
},
{
"epoch": 0.31,
"learning_rate": 0.0001384873949579832,
"loss": 0.0021,
"step": 550
},
{
"epoch": 0.31,
"learning_rate": 0.0001373669467787115,
"loss": 0.2876,
"step": 560
},
{
"epoch": 0.32,
"learning_rate": 0.00013624649859943978,
"loss": 0.0036,
"step": 570
},
{
"epoch": 0.32,
"learning_rate": 0.00013512605042016807,
"loss": 0.0026,
"step": 580
},
{
"epoch": 0.33,
"learning_rate": 0.00013400560224089634,
"loss": 0.0432,
"step": 590
},
{
"epoch": 0.34,
"learning_rate": 0.00013288515406162466,
"loss": 0.033,
"step": 600
},
{
"epoch": 0.34,
"learning_rate": 0.00013176470588235296,
"loss": 0.0048,
"step": 610
},
{
"epoch": 0.35,
"learning_rate": 0.00013064425770308125,
"loss": 0.0027,
"step": 620
},
{
"epoch": 0.35,
"learning_rate": 0.00012952380952380954,
"loss": 0.0021,
"step": 630
},
{
"epoch": 0.36,
"learning_rate": 0.00012840336134453784,
"loss": 0.0019,
"step": 640
},
{
"epoch": 0.36,
"learning_rate": 0.0001272829131652661,
"loss": 0.0018,
"step": 650
},
{
"epoch": 0.37,
"learning_rate": 0.0001261624649859944,
"loss": 0.0408,
"step": 660
},
{
"epoch": 0.38,
"learning_rate": 0.0001250420168067227,
"loss": 0.0105,
"step": 670
},
{
"epoch": 0.38,
"learning_rate": 0.000123921568627451,
"loss": 0.0023,
"step": 680
},
{
"epoch": 0.39,
"learning_rate": 0.00012280112044817928,
"loss": 0.0018,
"step": 690
},
{
"epoch": 0.39,
"learning_rate": 0.00012168067226890756,
"loss": 0.0247,
"step": 700
},
{
"epoch": 0.4,
"learning_rate": 0.00012056022408963586,
"loss": 0.0328,
"step": 710
},
{
"epoch": 0.4,
"learning_rate": 0.00011943977591036415,
"loss": 0.0081,
"step": 720
},
{
"epoch": 0.41,
"learning_rate": 0.00011831932773109244,
"loss": 0.0024,
"step": 730
},
{
"epoch": 0.41,
"learning_rate": 0.00011719887955182074,
"loss": 0.0018,
"step": 740
},
{
"epoch": 0.42,
"learning_rate": 0.00011607843137254903,
"loss": 0.0073,
"step": 750
},
{
"epoch": 0.43,
"learning_rate": 0.00011495798319327731,
"loss": 0.0015,
"step": 760
},
{
"epoch": 0.43,
"learning_rate": 0.0001138375350140056,
"loss": 0.0302,
"step": 770
},
{
"epoch": 0.44,
"learning_rate": 0.0001127170868347339,
"loss": 0.046,
"step": 780
},
{
"epoch": 0.44,
"learning_rate": 0.0001115966386554622,
"loss": 0.0306,
"step": 790
},
{
"epoch": 0.45,
"learning_rate": 0.00011047619047619049,
"loss": 0.0014,
"step": 800
},
{
"epoch": 0.45,
"learning_rate": 0.00010935574229691876,
"loss": 0.0149,
"step": 810
},
{
"epoch": 0.46,
"learning_rate": 0.00010823529411764706,
"loss": 0.0092,
"step": 820
},
{
"epoch": 0.46,
"learning_rate": 0.00010711484593837536,
"loss": 0.0069,
"step": 830
},
{
"epoch": 0.47,
"learning_rate": 0.00010599439775910365,
"loss": 0.0013,
"step": 840
},
{
"epoch": 0.48,
"learning_rate": 0.00010487394957983194,
"loss": 0.0017,
"step": 850
},
{
"epoch": 0.48,
"learning_rate": 0.00010375350140056024,
"loss": 0.0013,
"step": 860
},
{
"epoch": 0.49,
"learning_rate": 0.0001026330532212885,
"loss": 0.0175,
"step": 870
},
{
"epoch": 0.49,
"learning_rate": 0.0001015126050420168,
"loss": 0.0029,
"step": 880
},
{
"epoch": 0.5,
"learning_rate": 0.00010039215686274511,
"loss": 0.0023,
"step": 890
},
{
"epoch": 0.5,
"learning_rate": 9.92717086834734e-05,
"loss": 0.0018,
"step": 900
},
{
"epoch": 0.51,
"learning_rate": 9.815126050420168e-05,
"loss": 0.001,
"step": 910
},
{
"epoch": 0.52,
"learning_rate": 9.703081232492998e-05,
"loss": 0.0911,
"step": 920
},
{
"epoch": 0.52,
"learning_rate": 9.591036414565827e-05,
"loss": 0.0011,
"step": 930
},
{
"epoch": 0.53,
"learning_rate": 9.478991596638655e-05,
"loss": 0.0013,
"step": 940
},
{
"epoch": 0.53,
"learning_rate": 9.366946778711484e-05,
"loss": 0.0283,
"step": 950
},
{
"epoch": 0.54,
"learning_rate": 9.254901960784315e-05,
"loss": 0.0011,
"step": 960
},
{
"epoch": 0.54,
"learning_rate": 9.142857142857143e-05,
"loss": 0.0028,
"step": 970
},
{
"epoch": 0.55,
"learning_rate": 9.030812324929973e-05,
"loss": 0.0011,
"step": 980
},
{
"epoch": 0.55,
"learning_rate": 8.918767507002802e-05,
"loss": 0.001,
"step": 990
},
{
"epoch": 0.56,
"learning_rate": 8.80672268907563e-05,
"loss": 0.0186,
"step": 1000
},
{
"epoch": 0.56,
"eval_f1": 0.9990915697674418,
"eval_loss": 0.011647290550172329,
"eval_runtime": 81.5953,
"eval_samples_per_second": 69.244,
"eval_steps_per_second": 8.665,
"step": 1000
},
{
"epoch": 0.57,
"learning_rate": 8.69467787114846e-05,
"loss": 0.0352,
"step": 1010
},
{
"epoch": 0.57,
"learning_rate": 8.582633053221289e-05,
"loss": 0.0037,
"step": 1020
},
{
"epoch": 0.58,
"learning_rate": 8.470588235294118e-05,
"loss": 0.0014,
"step": 1030
},
{
"epoch": 0.58,
"learning_rate": 8.358543417366948e-05,
"loss": 0.005,
"step": 1040
},
{
"epoch": 0.59,
"learning_rate": 8.246498599439776e-05,
"loss": 0.001,
"step": 1050
},
{
"epoch": 0.59,
"learning_rate": 8.134453781512605e-05,
"loss": 0.0243,
"step": 1060
},
{
"epoch": 0.6,
"learning_rate": 8.022408963585435e-05,
"loss": 0.0012,
"step": 1070
},
{
"epoch": 0.61,
"learning_rate": 7.910364145658264e-05,
"loss": 0.0377,
"step": 1080
},
{
"epoch": 0.61,
"learning_rate": 7.798319327731093e-05,
"loss": 0.0011,
"step": 1090
},
{
"epoch": 0.62,
"learning_rate": 7.686274509803923e-05,
"loss": 0.0017,
"step": 1100
},
{
"epoch": 0.62,
"learning_rate": 7.574229691876751e-05,
"loss": 0.0061,
"step": 1110
},
{
"epoch": 0.63,
"learning_rate": 7.46218487394958e-05,
"loss": 0.0139,
"step": 1120
},
{
"epoch": 0.63,
"learning_rate": 7.35014005602241e-05,
"loss": 0.001,
"step": 1130
},
{
"epoch": 0.64,
"learning_rate": 7.238095238095238e-05,
"loss": 0.0008,
"step": 1140
},
{
"epoch": 0.64,
"learning_rate": 7.126050420168068e-05,
"loss": 0.0009,
"step": 1150
},
{
"epoch": 0.65,
"learning_rate": 7.014005602240896e-05,
"loss": 0.0024,
"step": 1160
},
{
"epoch": 0.66,
"learning_rate": 6.901960784313726e-05,
"loss": 0.0008,
"step": 1170
},
{
"epoch": 0.66,
"learning_rate": 6.789915966386555e-05,
"loss": 0.001,
"step": 1180
},
{
"epoch": 0.67,
"learning_rate": 6.677871148459383e-05,
"loss": 0.0008,
"step": 1190
},
{
"epoch": 0.67,
"learning_rate": 6.565826330532213e-05,
"loss": 0.0008,
"step": 1200
},
{
"epoch": 0.68,
"learning_rate": 6.453781512605043e-05,
"loss": 0.0353,
"step": 1210
},
{
"epoch": 0.68,
"learning_rate": 6.341736694677871e-05,
"loss": 0.0008,
"step": 1220
},
{
"epoch": 0.69,
"learning_rate": 6.229691876750701e-05,
"loss": 0.0011,
"step": 1230
},
{
"epoch": 0.69,
"learning_rate": 6.11764705882353e-05,
"loss": 0.0012,
"step": 1240
},
{
"epoch": 0.7,
"learning_rate": 6.005602240896358e-05,
"loss": 0.0012,
"step": 1250
},
{
"epoch": 0.71,
"learning_rate": 5.8935574229691884e-05,
"loss": 0.0008,
"step": 1260
},
{
"epoch": 0.71,
"learning_rate": 5.781512605042018e-05,
"loss": 0.0009,
"step": 1270
},
{
"epoch": 0.72,
"learning_rate": 5.669467787114846e-05,
"loss": 0.001,
"step": 1280
},
{
"epoch": 0.72,
"learning_rate": 5.557422969187676e-05,
"loss": 0.0007,
"step": 1290
},
{
"epoch": 0.73,
"learning_rate": 5.445378151260504e-05,
"loss": 0.001,
"step": 1300
},
{
"epoch": 0.73,
"learning_rate": 5.333333333333333e-05,
"loss": 0.0007,
"step": 1310
},
{
"epoch": 0.74,
"learning_rate": 5.221288515406163e-05,
"loss": 0.0007,
"step": 1320
},
{
"epoch": 0.75,
"learning_rate": 5.1092436974789914e-05,
"loss": 0.0007,
"step": 1330
},
{
"epoch": 0.75,
"learning_rate": 4.997198879551821e-05,
"loss": 0.001,
"step": 1340
},
{
"epoch": 0.76,
"learning_rate": 4.88515406162465e-05,
"loss": 0.0008,
"step": 1350
},
{
"epoch": 0.76,
"learning_rate": 4.7731092436974796e-05,
"loss": 0.0007,
"step": 1360
},
{
"epoch": 0.77,
"learning_rate": 4.6610644257703084e-05,
"loss": 0.0006,
"step": 1370
},
{
"epoch": 0.77,
"learning_rate": 4.549019607843137e-05,
"loss": 0.0008,
"step": 1380
},
{
"epoch": 0.78,
"learning_rate": 4.4369747899159665e-05,
"loss": 0.0006,
"step": 1390
},
{
"epoch": 0.78,
"learning_rate": 4.324929971988796e-05,
"loss": 0.0006,
"step": 1400
},
{
"epoch": 0.79,
"learning_rate": 4.2128851540616246e-05,
"loss": 0.0006,
"step": 1410
},
{
"epoch": 0.8,
"learning_rate": 4.100840336134454e-05,
"loss": 0.0192,
"step": 1420
},
{
"epoch": 0.8,
"learning_rate": 3.9887955182072834e-05,
"loss": 0.0006,
"step": 1430
},
{
"epoch": 0.81,
"learning_rate": 3.876750700280112e-05,
"loss": 0.0009,
"step": 1440
},
{
"epoch": 0.81,
"learning_rate": 3.7647058823529415e-05,
"loss": 0.0008,
"step": 1450
},
{
"epoch": 0.82,
"learning_rate": 3.65266106442577e-05,
"loss": 0.0006,
"step": 1460
},
{
"epoch": 0.82,
"learning_rate": 3.5406162464985996e-05,
"loss": 0.0006,
"step": 1470
},
{
"epoch": 0.83,
"learning_rate": 3.428571428571429e-05,
"loss": 0.0018,
"step": 1480
},
{
"epoch": 0.83,
"learning_rate": 3.316526610644258e-05,
"loss": 0.0006,
"step": 1490
},
{
"epoch": 0.84,
"learning_rate": 3.204481792717087e-05,
"loss": 0.0006,
"step": 1500
},
{
"epoch": 0.84,
"eval_f1": 0.9997273470871582,
"eval_loss": 0.004409410059452057,
"eval_runtime": 80.7573,
"eval_samples_per_second": 69.963,
"eval_steps_per_second": 8.755,
"step": 1500
},
{
"epoch": 0.85,
"learning_rate": 3.0924369747899166e-05,
"loss": 0.0006,
"step": 1510
},
{
"epoch": 0.85,
"learning_rate": 2.9803921568627453e-05,
"loss": 0.0011,
"step": 1520
},
{
"epoch": 0.86,
"learning_rate": 2.868347338935574e-05,
"loss": 0.0007,
"step": 1530
},
{
"epoch": 0.86,
"learning_rate": 2.7563025210084037e-05,
"loss": 0.0008,
"step": 1540
},
{
"epoch": 0.87,
"learning_rate": 2.6442577030812328e-05,
"loss": 0.0008,
"step": 1550
},
{
"epoch": 0.87,
"learning_rate": 2.5322128851540615e-05,
"loss": 0.0005,
"step": 1560
},
{
"epoch": 0.88,
"learning_rate": 2.420168067226891e-05,
"loss": 0.0005,
"step": 1570
},
{
"epoch": 0.89,
"learning_rate": 2.30812324929972e-05,
"loss": 0.0006,
"step": 1580
},
{
"epoch": 0.89,
"learning_rate": 2.196078431372549e-05,
"loss": 0.0005,
"step": 1590
},
{
"epoch": 0.9,
"learning_rate": 2.0840336134453785e-05,
"loss": 0.0006,
"step": 1600
},
{
"epoch": 0.9,
"learning_rate": 1.9719887955182072e-05,
"loss": 0.0006,
"step": 1610
},
{
"epoch": 0.91,
"learning_rate": 1.8599439775910366e-05,
"loss": 0.0005,
"step": 1620
},
{
"epoch": 0.91,
"learning_rate": 1.7478991596638656e-05,
"loss": 0.0005,
"step": 1630
},
{
"epoch": 0.92,
"learning_rate": 1.6358543417366947e-05,
"loss": 0.0005,
"step": 1640
},
{
"epoch": 0.92,
"learning_rate": 1.5238095238095241e-05,
"loss": 0.0005,
"step": 1650
},
{
"epoch": 0.93,
"learning_rate": 1.411764705882353e-05,
"loss": 0.0005,
"step": 1660
},
{
"epoch": 0.94,
"learning_rate": 1.2997198879551822e-05,
"loss": 0.0485,
"step": 1670
},
{
"epoch": 0.94,
"learning_rate": 1.1876750700280113e-05,
"loss": 0.0009,
"step": 1680
},
{
"epoch": 0.95,
"learning_rate": 1.0756302521008405e-05,
"loss": 0.0006,
"step": 1690
},
{
"epoch": 0.95,
"learning_rate": 9.635854341736696e-06,
"loss": 0.0005,
"step": 1700
},
{
"epoch": 0.96,
"learning_rate": 8.515406162464986e-06,
"loss": 0.0013,
"step": 1710
},
{
"epoch": 0.96,
"learning_rate": 7.394957983193278e-06,
"loss": 0.003,
"step": 1720
},
{
"epoch": 0.97,
"learning_rate": 6.274509803921569e-06,
"loss": 0.039,
"step": 1730
},
{
"epoch": 0.97,
"learning_rate": 5.15406162464986e-06,
"loss": 0.0005,
"step": 1740
},
{
"epoch": 0.98,
"learning_rate": 4.033613445378152e-06,
"loss": 0.0006,
"step": 1750
},
{
"epoch": 0.99,
"learning_rate": 2.9131652661064427e-06,
"loss": 0.0005,
"step": 1760
},
{
"epoch": 0.99,
"learning_rate": 1.7927170868347338e-06,
"loss": 0.0009,
"step": 1770
},
{
"epoch": 1.0,
"learning_rate": 6.722689075630253e-07,
"loss": 0.0007,
"step": 1780
},
{
"epoch": 1.0,
"step": 1785,
"total_flos": 2.2123963034966016e+18,
"train_loss": 0.019373888820827176,
"train_runtime": 828.3192,
"train_samples_per_second": 34.467,
"train_steps_per_second": 2.155
}
],
"max_steps": 1785,
"num_train_epochs": 1,
"total_flos": 2.2123963034966016e+18,
"trial_name": null,
"trial_params": null
}