mansee's picture
End of training
5376ff8
raw
history blame
20.9 kB
{
"best_metric": 0.7321109123434705,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-img_orientation/checkpoint-1258",
"epoch": 4.992050874403816,
"eval_steps": 500,
"global_step": 1570,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 3.1847133757961785e-06,
"loss": 1.4185,
"step": 10
},
{
"epoch": 0.06,
"learning_rate": 6.369426751592357e-06,
"loss": 1.41,
"step": 20
},
{
"epoch": 0.1,
"learning_rate": 9.554140127388536e-06,
"loss": 1.3794,
"step": 30
},
{
"epoch": 0.13,
"learning_rate": 1.2738853503184714e-05,
"loss": 1.3324,
"step": 40
},
{
"epoch": 0.16,
"learning_rate": 1.592356687898089e-05,
"loss": 1.3049,
"step": 50
},
{
"epoch": 0.19,
"learning_rate": 1.910828025477707e-05,
"loss": 1.2419,
"step": 60
},
{
"epoch": 0.22,
"learning_rate": 2.229299363057325e-05,
"loss": 1.1568,
"step": 70
},
{
"epoch": 0.25,
"learning_rate": 2.5477707006369428e-05,
"loss": 1.0499,
"step": 80
},
{
"epoch": 0.29,
"learning_rate": 2.8662420382165606e-05,
"loss": 0.9764,
"step": 90
},
{
"epoch": 0.32,
"learning_rate": 3.184713375796178e-05,
"loss": 0.8664,
"step": 100
},
{
"epoch": 0.35,
"learning_rate": 3.503184713375796e-05,
"loss": 0.814,
"step": 110
},
{
"epoch": 0.38,
"learning_rate": 3.821656050955414e-05,
"loss": 0.7859,
"step": 120
},
{
"epoch": 0.41,
"learning_rate": 4.1401273885350325e-05,
"loss": 0.7458,
"step": 130
},
{
"epoch": 0.45,
"learning_rate": 4.45859872611465e-05,
"loss": 0.782,
"step": 140
},
{
"epoch": 0.48,
"learning_rate": 4.777070063694268e-05,
"loss": 0.7633,
"step": 150
},
{
"epoch": 0.51,
"learning_rate": 4.9893842887473466e-05,
"loss": 0.7261,
"step": 160
},
{
"epoch": 0.54,
"learning_rate": 4.9539985845718334e-05,
"loss": 0.7286,
"step": 170
},
{
"epoch": 0.57,
"learning_rate": 4.91861288039632e-05,
"loss": 0.6912,
"step": 180
},
{
"epoch": 0.6,
"learning_rate": 4.883227176220807e-05,
"loss": 0.7093,
"step": 190
},
{
"epoch": 0.64,
"learning_rate": 4.8478414720452945e-05,
"loss": 0.6893,
"step": 200
},
{
"epoch": 0.67,
"learning_rate": 4.8124557678697806e-05,
"loss": 0.7246,
"step": 210
},
{
"epoch": 0.7,
"learning_rate": 4.777070063694268e-05,
"loss": 0.7175,
"step": 220
},
{
"epoch": 0.73,
"learning_rate": 4.741684359518755e-05,
"loss": 0.6927,
"step": 230
},
{
"epoch": 0.76,
"learning_rate": 4.706298655343242e-05,
"loss": 0.6424,
"step": 240
},
{
"epoch": 0.79,
"learning_rate": 4.6709129511677285e-05,
"loss": 0.6573,
"step": 250
},
{
"epoch": 0.83,
"learning_rate": 4.635527246992216e-05,
"loss": 0.6852,
"step": 260
},
{
"epoch": 0.86,
"learning_rate": 4.600141542816702e-05,
"loss": 0.6829,
"step": 270
},
{
"epoch": 0.89,
"learning_rate": 4.5647558386411895e-05,
"loss": 0.659,
"step": 280
},
{
"epoch": 0.92,
"learning_rate": 4.529370134465676e-05,
"loss": 0.6117,
"step": 290
},
{
"epoch": 0.95,
"learning_rate": 4.493984430290163e-05,
"loss": 0.6627,
"step": 300
},
{
"epoch": 0.99,
"learning_rate": 4.45859872611465e-05,
"loss": 0.6531,
"step": 310
},
{
"epoch": 1.0,
"eval_accuracy": 0.7043828264758497,
"eval_loss": 0.5172476768493652,
"eval_runtime": 33.3984,
"eval_samples_per_second": 133.899,
"eval_steps_per_second": 4.192,
"step": 314
},
{
"epoch": 1.02,
"learning_rate": 4.423213021939137e-05,
"loss": 0.6315,
"step": 320
},
{
"epoch": 1.05,
"learning_rate": 4.3878273177636235e-05,
"loss": 0.6119,
"step": 330
},
{
"epoch": 1.08,
"learning_rate": 4.352441613588111e-05,
"loss": 0.6335,
"step": 340
},
{
"epoch": 1.11,
"learning_rate": 4.317055909412597e-05,
"loss": 0.6371,
"step": 350
},
{
"epoch": 1.14,
"learning_rate": 4.2816702052370846e-05,
"loss": 0.5907,
"step": 360
},
{
"epoch": 1.18,
"learning_rate": 4.2462845010615714e-05,
"loss": 0.6202,
"step": 370
},
{
"epoch": 1.21,
"learning_rate": 4.210898796886058e-05,
"loss": 0.6195,
"step": 380
},
{
"epoch": 1.24,
"learning_rate": 4.175513092710545e-05,
"loss": 0.6165,
"step": 390
},
{
"epoch": 1.27,
"learning_rate": 4.1401273885350325e-05,
"loss": 0.6096,
"step": 400
},
{
"epoch": 1.3,
"learning_rate": 4.1047416843595186e-05,
"loss": 0.5847,
"step": 410
},
{
"epoch": 1.34,
"learning_rate": 4.069355980184006e-05,
"loss": 0.5893,
"step": 420
},
{
"epoch": 1.37,
"learning_rate": 4.033970276008493e-05,
"loss": 0.6183,
"step": 430
},
{
"epoch": 1.4,
"learning_rate": 3.99858457183298e-05,
"loss": 0.5857,
"step": 440
},
{
"epoch": 1.43,
"learning_rate": 3.9631988676574665e-05,
"loss": 0.5899,
"step": 450
},
{
"epoch": 1.46,
"learning_rate": 3.927813163481954e-05,
"loss": 0.6146,
"step": 460
},
{
"epoch": 1.49,
"learning_rate": 3.89242745930644e-05,
"loss": 0.5822,
"step": 470
},
{
"epoch": 1.53,
"learning_rate": 3.8570417551309275e-05,
"loss": 0.5783,
"step": 480
},
{
"epoch": 1.56,
"learning_rate": 3.821656050955414e-05,
"loss": 0.6289,
"step": 490
},
{
"epoch": 1.59,
"learning_rate": 3.786270346779901e-05,
"loss": 0.5934,
"step": 500
},
{
"epoch": 1.62,
"learning_rate": 3.750884642604388e-05,
"loss": 0.5836,
"step": 510
},
{
"epoch": 1.65,
"learning_rate": 3.715498938428875e-05,
"loss": 0.566,
"step": 520
},
{
"epoch": 1.69,
"learning_rate": 3.6801132342533615e-05,
"loss": 0.5933,
"step": 530
},
{
"epoch": 1.72,
"learning_rate": 3.644727530077849e-05,
"loss": 0.5699,
"step": 540
},
{
"epoch": 1.75,
"learning_rate": 3.609341825902335e-05,
"loss": 0.587,
"step": 550
},
{
"epoch": 1.78,
"learning_rate": 3.5739561217268226e-05,
"loss": 0.5909,
"step": 560
},
{
"epoch": 1.81,
"learning_rate": 3.5385704175513094e-05,
"loss": 0.5825,
"step": 570
},
{
"epoch": 1.84,
"learning_rate": 3.503184713375796e-05,
"loss": 0.5725,
"step": 580
},
{
"epoch": 1.88,
"learning_rate": 3.467799009200283e-05,
"loss": 0.5761,
"step": 590
},
{
"epoch": 1.91,
"learning_rate": 3.4324133050247705e-05,
"loss": 0.5869,
"step": 600
},
{
"epoch": 1.94,
"learning_rate": 3.3970276008492566e-05,
"loss": 0.623,
"step": 610
},
{
"epoch": 1.97,
"learning_rate": 3.361641896673744e-05,
"loss": 0.5774,
"step": 620
},
{
"epoch": 2.0,
"eval_accuracy": 0.7240608228980322,
"eval_loss": 0.461431086063385,
"eval_runtime": 32.7276,
"eval_samples_per_second": 136.643,
"eval_steps_per_second": 4.278,
"step": 629
},
{
"epoch": 2.0,
"learning_rate": 3.326256192498231e-05,
"loss": 0.5664,
"step": 630
},
{
"epoch": 2.03,
"learning_rate": 3.2908704883227177e-05,
"loss": 0.5354,
"step": 640
},
{
"epoch": 2.07,
"learning_rate": 3.2554847841472044e-05,
"loss": 0.5533,
"step": 650
},
{
"epoch": 2.1,
"learning_rate": 3.220099079971692e-05,
"loss": 0.5576,
"step": 660
},
{
"epoch": 2.13,
"learning_rate": 3.184713375796178e-05,
"loss": 0.5637,
"step": 670
},
{
"epoch": 2.16,
"learning_rate": 3.1493276716206655e-05,
"loss": 0.5502,
"step": 680
},
{
"epoch": 2.19,
"learning_rate": 3.113941967445152e-05,
"loss": 0.5451,
"step": 690
},
{
"epoch": 2.23,
"learning_rate": 3.078556263269639e-05,
"loss": 0.5996,
"step": 700
},
{
"epoch": 2.26,
"learning_rate": 3.0431705590941263e-05,
"loss": 0.5754,
"step": 710
},
{
"epoch": 2.29,
"learning_rate": 3.0077848549186134e-05,
"loss": 0.5321,
"step": 720
},
{
"epoch": 2.32,
"learning_rate": 2.9723991507431e-05,
"loss": 0.5443,
"step": 730
},
{
"epoch": 2.35,
"learning_rate": 2.937013446567587e-05,
"loss": 0.5471,
"step": 740
},
{
"epoch": 2.38,
"learning_rate": 2.9016277423920734e-05,
"loss": 0.5147,
"step": 750
},
{
"epoch": 2.42,
"learning_rate": 2.8662420382165606e-05,
"loss": 0.556,
"step": 760
},
{
"epoch": 2.45,
"learning_rate": 2.8308563340410477e-05,
"loss": 0.5613,
"step": 770
},
{
"epoch": 2.48,
"learning_rate": 2.7954706298655342e-05,
"loss": 0.5748,
"step": 780
},
{
"epoch": 2.51,
"learning_rate": 2.7600849256900213e-05,
"loss": 0.5698,
"step": 790
},
{
"epoch": 2.54,
"learning_rate": 2.7246992215145084e-05,
"loss": 0.561,
"step": 800
},
{
"epoch": 2.58,
"learning_rate": 2.689313517338995e-05,
"loss": 0.5444,
"step": 810
},
{
"epoch": 2.61,
"learning_rate": 2.653927813163482e-05,
"loss": 0.5156,
"step": 820
},
{
"epoch": 2.64,
"learning_rate": 2.6185421089879692e-05,
"loss": 0.5358,
"step": 830
},
{
"epoch": 2.67,
"learning_rate": 2.5831564048124556e-05,
"loss": 0.5156,
"step": 840
},
{
"epoch": 2.7,
"learning_rate": 2.5477707006369428e-05,
"loss": 0.5386,
"step": 850
},
{
"epoch": 2.73,
"learning_rate": 2.51238499646143e-05,
"loss": 0.5513,
"step": 860
},
{
"epoch": 2.77,
"learning_rate": 2.4769992922859167e-05,
"loss": 0.5217,
"step": 870
},
{
"epoch": 2.8,
"learning_rate": 2.4416135881104035e-05,
"loss": 0.5621,
"step": 880
},
{
"epoch": 2.83,
"learning_rate": 2.4062278839348903e-05,
"loss": 0.5457,
"step": 890
},
{
"epoch": 2.86,
"learning_rate": 2.3708421797593774e-05,
"loss": 0.5172,
"step": 900
},
{
"epoch": 2.89,
"learning_rate": 2.3354564755838642e-05,
"loss": 0.5598,
"step": 910
},
{
"epoch": 2.93,
"learning_rate": 2.300070771408351e-05,
"loss": 0.5391,
"step": 920
},
{
"epoch": 2.96,
"learning_rate": 2.264685067232838e-05,
"loss": 0.5284,
"step": 930
},
{
"epoch": 2.99,
"learning_rate": 2.229299363057325e-05,
"loss": 0.5496,
"step": 940
},
{
"epoch": 3.0,
"eval_accuracy": 0.7298747763864043,
"eval_loss": 0.4499094486236572,
"eval_runtime": 32.2417,
"eval_samples_per_second": 138.702,
"eval_steps_per_second": 4.342,
"step": 943
},
{
"epoch": 3.02,
"learning_rate": 2.1939136588818118e-05,
"loss": 0.5027,
"step": 950
},
{
"epoch": 3.05,
"learning_rate": 2.1585279547062986e-05,
"loss": 0.5285,
"step": 960
},
{
"epoch": 3.08,
"learning_rate": 2.1231422505307857e-05,
"loss": 0.4934,
"step": 970
},
{
"epoch": 3.12,
"learning_rate": 2.0877565463552725e-05,
"loss": 0.5001,
"step": 980
},
{
"epoch": 3.15,
"learning_rate": 2.0523708421797593e-05,
"loss": 0.5378,
"step": 990
},
{
"epoch": 3.18,
"learning_rate": 2.0169851380042464e-05,
"loss": 0.4971,
"step": 1000
},
{
"epoch": 3.21,
"learning_rate": 1.9815994338287332e-05,
"loss": 0.5028,
"step": 1010
},
{
"epoch": 3.24,
"learning_rate": 1.94621372965322e-05,
"loss": 0.5353,
"step": 1020
},
{
"epoch": 3.28,
"learning_rate": 1.910828025477707e-05,
"loss": 0.5243,
"step": 1030
},
{
"epoch": 3.31,
"learning_rate": 1.875442321302194e-05,
"loss": 0.5302,
"step": 1040
},
{
"epoch": 3.34,
"learning_rate": 1.8400566171266808e-05,
"loss": 0.5101,
"step": 1050
},
{
"epoch": 3.37,
"learning_rate": 1.8046709129511676e-05,
"loss": 0.5032,
"step": 1060
},
{
"epoch": 3.4,
"learning_rate": 1.7692852087756547e-05,
"loss": 0.5075,
"step": 1070
},
{
"epoch": 3.43,
"learning_rate": 1.7338995046001415e-05,
"loss": 0.5148,
"step": 1080
},
{
"epoch": 3.47,
"learning_rate": 1.6985138004246283e-05,
"loss": 0.5152,
"step": 1090
},
{
"epoch": 3.5,
"learning_rate": 1.6631280962491154e-05,
"loss": 0.5222,
"step": 1100
},
{
"epoch": 3.53,
"learning_rate": 1.6277423920736022e-05,
"loss": 0.4955,
"step": 1110
},
{
"epoch": 3.56,
"learning_rate": 1.592356687898089e-05,
"loss": 0.5029,
"step": 1120
},
{
"epoch": 3.59,
"learning_rate": 1.556970983722576e-05,
"loss": 0.516,
"step": 1130
},
{
"epoch": 3.62,
"learning_rate": 1.5215852795470631e-05,
"loss": 0.5105,
"step": 1140
},
{
"epoch": 3.66,
"learning_rate": 1.48619957537155e-05,
"loss": 0.5064,
"step": 1150
},
{
"epoch": 3.69,
"learning_rate": 1.4508138711960367e-05,
"loss": 0.5082,
"step": 1160
},
{
"epoch": 3.72,
"learning_rate": 1.4154281670205239e-05,
"loss": 0.4877,
"step": 1170
},
{
"epoch": 3.75,
"learning_rate": 1.3800424628450107e-05,
"loss": 0.5252,
"step": 1180
},
{
"epoch": 3.78,
"learning_rate": 1.3446567586694975e-05,
"loss": 0.5415,
"step": 1190
},
{
"epoch": 3.82,
"learning_rate": 1.3092710544939846e-05,
"loss": 0.4911,
"step": 1200
},
{
"epoch": 3.85,
"learning_rate": 1.2738853503184714e-05,
"loss": 0.5024,
"step": 1210
},
{
"epoch": 3.88,
"learning_rate": 1.2384996461429584e-05,
"loss": 0.5201,
"step": 1220
},
{
"epoch": 3.91,
"learning_rate": 1.2031139419674452e-05,
"loss": 0.5291,
"step": 1230
},
{
"epoch": 3.94,
"learning_rate": 1.1677282377919321e-05,
"loss": 0.5271,
"step": 1240
},
{
"epoch": 3.97,
"learning_rate": 1.132342533616419e-05,
"loss": 0.5014,
"step": 1250
},
{
"epoch": 4.0,
"eval_accuracy": 0.7321109123434705,
"eval_loss": 0.45054975152015686,
"eval_runtime": 31.9873,
"eval_samples_per_second": 139.805,
"eval_steps_per_second": 4.377,
"step": 1258
},
{
"epoch": 4.01,
"learning_rate": 1.0969568294409059e-05,
"loss": 0.5164,
"step": 1260
},
{
"epoch": 4.04,
"learning_rate": 1.0615711252653929e-05,
"loss": 0.4968,
"step": 1270
},
{
"epoch": 4.07,
"learning_rate": 1.0261854210898796e-05,
"loss": 0.4781,
"step": 1280
},
{
"epoch": 4.1,
"learning_rate": 9.907997169143666e-06,
"loss": 0.4935,
"step": 1290
},
{
"epoch": 4.13,
"learning_rate": 9.554140127388536e-06,
"loss": 0.455,
"step": 1300
},
{
"epoch": 4.17,
"learning_rate": 9.200283085633404e-06,
"loss": 0.5263,
"step": 1310
},
{
"epoch": 4.2,
"learning_rate": 8.846426043878273e-06,
"loss": 0.4888,
"step": 1320
},
{
"epoch": 4.23,
"learning_rate": 8.492569002123141e-06,
"loss": 0.4853,
"step": 1330
},
{
"epoch": 4.26,
"learning_rate": 8.138711960368011e-06,
"loss": 0.4744,
"step": 1340
},
{
"epoch": 4.29,
"learning_rate": 7.78485491861288e-06,
"loss": 0.4537,
"step": 1350
},
{
"epoch": 4.32,
"learning_rate": 7.43099787685775e-06,
"loss": 0.5045,
"step": 1360
},
{
"epoch": 4.36,
"learning_rate": 7.077140835102619e-06,
"loss": 0.4707,
"step": 1370
},
{
"epoch": 4.39,
"learning_rate": 6.723283793347487e-06,
"loss": 0.4776,
"step": 1380
},
{
"epoch": 4.42,
"learning_rate": 6.369426751592357e-06,
"loss": 0.5119,
"step": 1390
},
{
"epoch": 4.45,
"learning_rate": 6.015569709837226e-06,
"loss": 0.5521,
"step": 1400
},
{
"epoch": 4.48,
"learning_rate": 5.661712668082095e-06,
"loss": 0.4871,
"step": 1410
},
{
"epoch": 4.52,
"learning_rate": 5.307855626326964e-06,
"loss": 0.5006,
"step": 1420
},
{
"epoch": 4.55,
"learning_rate": 4.953998584571833e-06,
"loss": 0.4733,
"step": 1430
},
{
"epoch": 4.58,
"learning_rate": 4.600141542816702e-06,
"loss": 0.4942,
"step": 1440
},
{
"epoch": 4.61,
"learning_rate": 4.246284501061571e-06,
"loss": 0.4806,
"step": 1450
},
{
"epoch": 4.64,
"learning_rate": 3.89242745930644e-06,
"loss": 0.5122,
"step": 1460
},
{
"epoch": 4.67,
"learning_rate": 3.5385704175513096e-06,
"loss": 0.4843,
"step": 1470
},
{
"epoch": 4.71,
"learning_rate": 3.1847133757961785e-06,
"loss": 0.4757,
"step": 1480
},
{
"epoch": 4.74,
"learning_rate": 2.8308563340410473e-06,
"loss": 0.4717,
"step": 1490
},
{
"epoch": 4.77,
"learning_rate": 2.4769992922859165e-06,
"loss": 0.5049,
"step": 1500
},
{
"epoch": 4.8,
"learning_rate": 2.1231422505307854e-06,
"loss": 0.4666,
"step": 1510
},
{
"epoch": 4.83,
"learning_rate": 1.7692852087756548e-06,
"loss": 0.4781,
"step": 1520
},
{
"epoch": 4.86,
"learning_rate": 1.4154281670205236e-06,
"loss": 0.485,
"step": 1530
},
{
"epoch": 4.9,
"learning_rate": 1.0615711252653927e-06,
"loss": 0.5166,
"step": 1540
},
{
"epoch": 4.93,
"learning_rate": 7.077140835102618e-07,
"loss": 0.5056,
"step": 1550
},
{
"epoch": 4.96,
"learning_rate": 3.538570417551309e-07,
"loss": 0.4843,
"step": 1560
},
{
"epoch": 4.99,
"learning_rate": 0.0,
"loss": 0.4746,
"step": 1570
},
{
"epoch": 4.99,
"eval_accuracy": 0.7294275491949911,
"eval_loss": 0.45043760538101196,
"eval_runtime": 33.0157,
"eval_samples_per_second": 135.451,
"eval_steps_per_second": 4.24,
"step": 1570
},
{
"epoch": 4.99,
"step": 1570,
"total_flos": 4.994244969949348e+18,
"train_loss": 0.6026059660941931,
"train_runtime": 3290.1425,
"train_samples_per_second": 61.16,
"train_steps_per_second": 0.477
}
],
"logging_steps": 10,
"max_steps": 1570,
"num_train_epochs": 5,
"save_steps": 500,
"total_flos": 4.994244969949348e+18,
"trial_name": null,
"trial_params": null
}