vit-base-cats-vs-dogs / trainer_state.json
yourusername's picture
add model
a1ed873
{
"best_metric": 0.020224373787641525,
"best_model_checkpoint": "./cats_vs_dogs_outputs/checkpoint-1555",
"epoch": 5.0,
"global_step": 1555,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.03,
"learning_rate": 0.00019871382636655948,
"loss": 0.3131,
"step": 10
},
{
"epoch": 0.06,
"learning_rate": 0.00019742765273311899,
"loss": 0.1692,
"step": 20
},
{
"epoch": 0.1,
"learning_rate": 0.00019614147909967846,
"loss": 0.0963,
"step": 30
},
{
"epoch": 0.13,
"learning_rate": 0.00019485530546623796,
"loss": 0.0968,
"step": 40
},
{
"epoch": 0.16,
"learning_rate": 0.00019356913183279743,
"loss": 0.1045,
"step": 50
},
{
"epoch": 0.19,
"learning_rate": 0.00019228295819935694,
"loss": 0.0935,
"step": 60
},
{
"epoch": 0.23,
"learning_rate": 0.0001909967845659164,
"loss": 0.1173,
"step": 70
},
{
"epoch": 0.26,
"learning_rate": 0.00018971061093247588,
"loss": 0.0906,
"step": 80
},
{
"epoch": 0.29,
"learning_rate": 0.00018842443729903539,
"loss": 0.1101,
"step": 90
},
{
"epoch": 0.32,
"learning_rate": 0.00018713826366559486,
"loss": 0.1224,
"step": 100
},
{
"epoch": 0.35,
"learning_rate": 0.00018585209003215436,
"loss": 0.0655,
"step": 110
},
{
"epoch": 0.39,
"learning_rate": 0.00018456591639871384,
"loss": 0.1049,
"step": 120
},
{
"epoch": 0.42,
"learning_rate": 0.0001832797427652733,
"loss": 0.1104,
"step": 130
},
{
"epoch": 0.45,
"learning_rate": 0.0001819935691318328,
"loss": 0.0567,
"step": 140
},
{
"epoch": 0.48,
"learning_rate": 0.00018070739549839229,
"loss": 0.0609,
"step": 150
},
{
"epoch": 0.51,
"learning_rate": 0.0001794212218649518,
"loss": 0.0684,
"step": 160
},
{
"epoch": 0.55,
"learning_rate": 0.00017813504823151126,
"loss": 0.086,
"step": 170
},
{
"epoch": 0.58,
"learning_rate": 0.00017684887459807076,
"loss": 0.0708,
"step": 180
},
{
"epoch": 0.61,
"learning_rate": 0.00017556270096463024,
"loss": 0.0693,
"step": 190
},
{
"epoch": 0.64,
"learning_rate": 0.0001742765273311897,
"loss": 0.0814,
"step": 200
},
{
"epoch": 0.68,
"learning_rate": 0.0001729903536977492,
"loss": 0.1044,
"step": 210
},
{
"epoch": 0.71,
"learning_rate": 0.0001717041800643087,
"loss": 0.0896,
"step": 220
},
{
"epoch": 0.74,
"learning_rate": 0.0001704180064308682,
"loss": 0.1199,
"step": 230
},
{
"epoch": 0.77,
"learning_rate": 0.00016913183279742766,
"loss": 0.0773,
"step": 240
},
{
"epoch": 0.8,
"learning_rate": 0.00016784565916398716,
"loss": 0.0845,
"step": 250
},
{
"epoch": 0.84,
"learning_rate": 0.00016655948553054664,
"loss": 0.087,
"step": 260
},
{
"epoch": 0.87,
"learning_rate": 0.0001652733118971061,
"loss": 0.0836,
"step": 270
},
{
"epoch": 0.9,
"learning_rate": 0.0001639871382636656,
"loss": 0.0781,
"step": 280
},
{
"epoch": 0.93,
"learning_rate": 0.0001627009646302251,
"loss": 0.0764,
"step": 290
},
{
"epoch": 0.96,
"learning_rate": 0.0001614147909967846,
"loss": 0.0576,
"step": 300
},
{
"epoch": 1.0,
"learning_rate": 0.00016012861736334406,
"loss": 0.064,
"step": 310
},
{
"epoch": 1.0,
"eval_accuracy": 0.9849088838268792,
"eval_loss": 0.048329003155231476,
"eval_runtime": 24.3841,
"eval_samples_per_second": 144.028,
"eval_steps_per_second": 2.256,
"step": 311
},
{
"epoch": 1.03,
"learning_rate": 0.00015884244372990354,
"loss": 0.0519,
"step": 320
},
{
"epoch": 1.06,
"learning_rate": 0.00015755627009646304,
"loss": 0.0666,
"step": 330
},
{
"epoch": 1.09,
"learning_rate": 0.0001562700964630225,
"loss": 0.0706,
"step": 340
},
{
"epoch": 1.13,
"learning_rate": 0.00015498392282958201,
"loss": 0.0632,
"step": 350
},
{
"epoch": 1.16,
"learning_rate": 0.0001536977491961415,
"loss": 0.0836,
"step": 360
},
{
"epoch": 1.19,
"learning_rate": 0.000152411575562701,
"loss": 0.0873,
"step": 370
},
{
"epoch": 1.22,
"learning_rate": 0.00015112540192926046,
"loss": 0.0711,
"step": 380
},
{
"epoch": 1.25,
"learning_rate": 0.00014983922829581994,
"loss": 0.0694,
"step": 390
},
{
"epoch": 1.29,
"learning_rate": 0.00014855305466237944,
"loss": 0.0803,
"step": 400
},
{
"epoch": 1.32,
"learning_rate": 0.00014726688102893891,
"loss": 0.0544,
"step": 410
},
{
"epoch": 1.35,
"learning_rate": 0.00014598070739549841,
"loss": 0.0673,
"step": 420
},
{
"epoch": 1.38,
"learning_rate": 0.0001446945337620579,
"loss": 0.0751,
"step": 430
},
{
"epoch": 1.41,
"learning_rate": 0.0001434083601286174,
"loss": 0.0567,
"step": 440
},
{
"epoch": 1.45,
"learning_rate": 0.00014212218649517686,
"loss": 0.044,
"step": 450
},
{
"epoch": 1.48,
"learning_rate": 0.00014083601286173634,
"loss": 0.0577,
"step": 460
},
{
"epoch": 1.51,
"learning_rate": 0.00013954983922829584,
"loss": 0.0451,
"step": 470
},
{
"epoch": 1.54,
"learning_rate": 0.00013826366559485531,
"loss": 0.0803,
"step": 480
},
{
"epoch": 1.58,
"learning_rate": 0.00013697749196141482,
"loss": 0.0481,
"step": 490
},
{
"epoch": 1.61,
"learning_rate": 0.0001356913183279743,
"loss": 0.0716,
"step": 500
},
{
"epoch": 1.64,
"learning_rate": 0.00013440514469453376,
"loss": 0.0568,
"step": 510
},
{
"epoch": 1.67,
"learning_rate": 0.00013311897106109327,
"loss": 0.0724,
"step": 520
},
{
"epoch": 1.7,
"learning_rate": 0.00013183279742765274,
"loss": 0.072,
"step": 530
},
{
"epoch": 1.74,
"learning_rate": 0.00013054662379421224,
"loss": 0.0456,
"step": 540
},
{
"epoch": 1.77,
"learning_rate": 0.00012926045016077172,
"loss": 0.0548,
"step": 550
},
{
"epoch": 1.8,
"learning_rate": 0.00012797427652733122,
"loss": 0.0566,
"step": 560
},
{
"epoch": 1.83,
"learning_rate": 0.0001266881028938907,
"loss": 0.0665,
"step": 570
},
{
"epoch": 1.86,
"learning_rate": 0.00012540192926045017,
"loss": 0.0553,
"step": 580
},
{
"epoch": 1.9,
"learning_rate": 0.00012411575562700967,
"loss": 0.058,
"step": 590
},
{
"epoch": 1.93,
"learning_rate": 0.00012282958199356914,
"loss": 0.0532,
"step": 600
},
{
"epoch": 1.96,
"learning_rate": 0.00012154340836012863,
"loss": 0.0381,
"step": 610
},
{
"epoch": 1.99,
"learning_rate": 0.0001202572347266881,
"loss": 0.0622,
"step": 620
},
{
"epoch": 2.0,
"eval_accuracy": 0.9903189066059226,
"eval_loss": 0.027488160878419876,
"eval_runtime": 24.3864,
"eval_samples_per_second": 144.015,
"eval_steps_per_second": 2.255,
"step": 622
},
{
"epoch": 2.03,
"learning_rate": 0.0001189710610932476,
"loss": 0.0454,
"step": 630
},
{
"epoch": 2.06,
"learning_rate": 0.00011768488745980708,
"loss": 0.0383,
"step": 640
},
{
"epoch": 2.09,
"learning_rate": 0.00011639871382636655,
"loss": 0.0545,
"step": 650
},
{
"epoch": 2.12,
"learning_rate": 0.00011511254019292605,
"loss": 0.063,
"step": 660
},
{
"epoch": 2.15,
"learning_rate": 0.00011382636655948553,
"loss": 0.0717,
"step": 670
},
{
"epoch": 2.19,
"learning_rate": 0.00011254019292604503,
"loss": 0.0545,
"step": 680
},
{
"epoch": 2.22,
"learning_rate": 0.0001112540192926045,
"loss": 0.0576,
"step": 690
},
{
"epoch": 2.25,
"learning_rate": 0.00010996784565916398,
"loss": 0.0656,
"step": 700
},
{
"epoch": 2.28,
"learning_rate": 0.00010868167202572348,
"loss": 0.0762,
"step": 710
},
{
"epoch": 2.32,
"learning_rate": 0.00010739549839228295,
"loss": 0.0555,
"step": 720
},
{
"epoch": 2.35,
"learning_rate": 0.00010610932475884245,
"loss": 0.0492,
"step": 730
},
{
"epoch": 2.38,
"learning_rate": 0.00010482315112540193,
"loss": 0.042,
"step": 740
},
{
"epoch": 2.41,
"learning_rate": 0.00010353697749196143,
"loss": 0.0361,
"step": 750
},
{
"epoch": 2.44,
"learning_rate": 0.0001022508038585209,
"loss": 0.0522,
"step": 760
},
{
"epoch": 2.48,
"learning_rate": 0.00010096463022508038,
"loss": 0.0577,
"step": 770
},
{
"epoch": 2.51,
"learning_rate": 9.967845659163988e-05,
"loss": 0.0339,
"step": 780
},
{
"epoch": 2.54,
"learning_rate": 9.839228295819937e-05,
"loss": 0.0439,
"step": 790
},
{
"epoch": 2.57,
"learning_rate": 9.710610932475884e-05,
"loss": 0.0422,
"step": 800
},
{
"epoch": 2.6,
"learning_rate": 9.581993569131833e-05,
"loss": 0.03,
"step": 810
},
{
"epoch": 2.64,
"learning_rate": 9.453376205787782e-05,
"loss": 0.0637,
"step": 820
},
{
"epoch": 2.67,
"learning_rate": 9.32475884244373e-05,
"loss": 0.0629,
"step": 830
},
{
"epoch": 2.7,
"learning_rate": 9.19614147909968e-05,
"loss": 0.0371,
"step": 840
},
{
"epoch": 2.73,
"learning_rate": 9.067524115755628e-05,
"loss": 0.0419,
"step": 850
},
{
"epoch": 2.77,
"learning_rate": 8.938906752411576e-05,
"loss": 0.0479,
"step": 860
},
{
"epoch": 2.8,
"learning_rate": 8.810289389067524e-05,
"loss": 0.0392,
"step": 870
},
{
"epoch": 2.83,
"learning_rate": 8.681672025723473e-05,
"loss": 0.0383,
"step": 880
},
{
"epoch": 2.86,
"learning_rate": 8.553054662379422e-05,
"loss": 0.0449,
"step": 890
},
{
"epoch": 2.89,
"learning_rate": 8.42443729903537e-05,
"loss": 0.0481,
"step": 900
},
{
"epoch": 2.93,
"learning_rate": 8.29581993569132e-05,
"loss": 0.0292,
"step": 910
},
{
"epoch": 2.96,
"learning_rate": 8.167202572347268e-05,
"loss": 0.04,
"step": 920
},
{
"epoch": 2.99,
"learning_rate": 8.038585209003216e-05,
"loss": 0.0366,
"step": 930
},
{
"epoch": 3.0,
"eval_accuracy": 0.9917425968109339,
"eval_loss": 0.026212546974420547,
"eval_runtime": 24.1897,
"eval_samples_per_second": 145.186,
"eval_steps_per_second": 2.274,
"step": 933
},
{
"epoch": 3.02,
"learning_rate": 7.909967845659164e-05,
"loss": 0.0577,
"step": 940
},
{
"epoch": 3.05,
"learning_rate": 7.781350482315113e-05,
"loss": 0.049,
"step": 950
},
{
"epoch": 3.09,
"learning_rate": 7.652733118971062e-05,
"loss": 0.0366,
"step": 960
},
{
"epoch": 3.12,
"learning_rate": 7.524115755627011e-05,
"loss": 0.0286,
"step": 970
},
{
"epoch": 3.15,
"learning_rate": 7.39549839228296e-05,
"loss": 0.038,
"step": 980
},
{
"epoch": 3.18,
"learning_rate": 7.266881028938907e-05,
"loss": 0.0386,
"step": 990
},
{
"epoch": 3.22,
"learning_rate": 7.138263665594856e-05,
"loss": 0.0287,
"step": 1000
},
{
"epoch": 3.25,
"learning_rate": 7.009646302250804e-05,
"loss": 0.0284,
"step": 1010
},
{
"epoch": 3.28,
"learning_rate": 6.881028938906753e-05,
"loss": 0.0267,
"step": 1020
},
{
"epoch": 3.31,
"learning_rate": 6.752411575562702e-05,
"loss": 0.0305,
"step": 1030
},
{
"epoch": 3.34,
"learning_rate": 6.623794212218651e-05,
"loss": 0.0438,
"step": 1040
},
{
"epoch": 3.38,
"learning_rate": 6.495176848874598e-05,
"loss": 0.0265,
"step": 1050
},
{
"epoch": 3.41,
"learning_rate": 6.366559485530547e-05,
"loss": 0.0346,
"step": 1060
},
{
"epoch": 3.44,
"learning_rate": 6.237942122186496e-05,
"loss": 0.0468,
"step": 1070
},
{
"epoch": 3.47,
"learning_rate": 6.109324758842445e-05,
"loss": 0.0343,
"step": 1080
},
{
"epoch": 3.5,
"learning_rate": 5.980707395498393e-05,
"loss": 0.0664,
"step": 1090
},
{
"epoch": 3.54,
"learning_rate": 5.8520900321543414e-05,
"loss": 0.0587,
"step": 1100
},
{
"epoch": 3.57,
"learning_rate": 5.72347266881029e-05,
"loss": 0.0322,
"step": 1110
},
{
"epoch": 3.6,
"learning_rate": 5.5948553054662377e-05,
"loss": 0.0323,
"step": 1120
},
{
"epoch": 3.63,
"learning_rate": 5.4662379421221864e-05,
"loss": 0.033,
"step": 1130
},
{
"epoch": 3.67,
"learning_rate": 5.337620578778135e-05,
"loss": 0.0343,
"step": 1140
},
{
"epoch": 3.7,
"learning_rate": 5.209003215434084e-05,
"loss": 0.0463,
"step": 1150
},
{
"epoch": 3.73,
"learning_rate": 5.080385852090033e-05,
"loss": 0.0249,
"step": 1160
},
{
"epoch": 3.76,
"learning_rate": 4.951768488745981e-05,
"loss": 0.0391,
"step": 1170
},
{
"epoch": 3.79,
"learning_rate": 4.8231511254019296e-05,
"loss": 0.0371,
"step": 1180
},
{
"epoch": 3.83,
"learning_rate": 4.6945337620578784e-05,
"loss": 0.0288,
"step": 1190
},
{
"epoch": 3.86,
"learning_rate": 4.5659163987138265e-05,
"loss": 0.0239,
"step": 1200
},
{
"epoch": 3.89,
"learning_rate": 4.437299035369775e-05,
"loss": 0.0154,
"step": 1210
},
{
"epoch": 3.92,
"learning_rate": 4.308681672025724e-05,
"loss": 0.0213,
"step": 1220
},
{
"epoch": 3.95,
"learning_rate": 4.180064308681672e-05,
"loss": 0.0543,
"step": 1230
},
{
"epoch": 3.99,
"learning_rate": 4.051446945337621e-05,
"loss": 0.0294,
"step": 1240
},
{
"epoch": 4.0,
"eval_accuracy": 0.9931662870159453,
"eval_loss": 0.02194945700466633,
"eval_runtime": 24.0412,
"eval_samples_per_second": 146.083,
"eval_steps_per_second": 2.288,
"step": 1244
},
{
"epoch": 4.02,
"learning_rate": 3.92282958199357e-05,
"loss": 0.0325,
"step": 1250
},
{
"epoch": 4.05,
"learning_rate": 3.794212218649518e-05,
"loss": 0.02,
"step": 1260
},
{
"epoch": 4.08,
"learning_rate": 3.6655948553054666e-05,
"loss": 0.0284,
"step": 1270
},
{
"epoch": 4.12,
"learning_rate": 3.5369774919614154e-05,
"loss": 0.0266,
"step": 1280
},
{
"epoch": 4.15,
"learning_rate": 3.4083601286173635e-05,
"loss": 0.0232,
"step": 1290
},
{
"epoch": 4.18,
"learning_rate": 3.279742765273312e-05,
"loss": 0.0193,
"step": 1300
},
{
"epoch": 4.21,
"learning_rate": 3.151125401929261e-05,
"loss": 0.0286,
"step": 1310
},
{
"epoch": 4.24,
"learning_rate": 3.0225080385852088e-05,
"loss": 0.0464,
"step": 1320
},
{
"epoch": 4.28,
"learning_rate": 2.8938906752411576e-05,
"loss": 0.0234,
"step": 1330
},
{
"epoch": 4.31,
"learning_rate": 2.7652733118971064e-05,
"loss": 0.0331,
"step": 1340
},
{
"epoch": 4.34,
"learning_rate": 2.6366559485530545e-05,
"loss": 0.0243,
"step": 1350
},
{
"epoch": 4.37,
"learning_rate": 2.5080385852090032e-05,
"loss": 0.0161,
"step": 1360
},
{
"epoch": 4.41,
"learning_rate": 2.379421221864952e-05,
"loss": 0.0228,
"step": 1370
},
{
"epoch": 4.44,
"learning_rate": 2.2508038585209005e-05,
"loss": 0.0127,
"step": 1380
},
{
"epoch": 4.47,
"learning_rate": 2.122186495176849e-05,
"loss": 0.0281,
"step": 1390
},
{
"epoch": 4.5,
"learning_rate": 1.9935691318327977e-05,
"loss": 0.0369,
"step": 1400
},
{
"epoch": 4.53,
"learning_rate": 1.864951768488746e-05,
"loss": 0.0205,
"step": 1410
},
{
"epoch": 4.57,
"learning_rate": 1.736334405144695e-05,
"loss": 0.0263,
"step": 1420
},
{
"epoch": 4.6,
"learning_rate": 1.6077170418006433e-05,
"loss": 0.0106,
"step": 1430
},
{
"epoch": 4.63,
"learning_rate": 1.4790996784565916e-05,
"loss": 0.0215,
"step": 1440
},
{
"epoch": 4.66,
"learning_rate": 1.3504823151125404e-05,
"loss": 0.0251,
"step": 1450
},
{
"epoch": 4.69,
"learning_rate": 1.2218649517684888e-05,
"loss": 0.0177,
"step": 1460
},
{
"epoch": 4.73,
"learning_rate": 1.0932475884244374e-05,
"loss": 0.0224,
"step": 1470
},
{
"epoch": 4.76,
"learning_rate": 9.646302250803859e-06,
"loss": 0.0242,
"step": 1480
},
{
"epoch": 4.79,
"learning_rate": 8.360128617363345e-06,
"loss": 0.0188,
"step": 1490
},
{
"epoch": 4.82,
"learning_rate": 7.07395498392283e-06,
"loss": 0.0226,
"step": 1500
},
{
"epoch": 4.86,
"learning_rate": 5.787781350482315e-06,
"loss": 0.0417,
"step": 1510
},
{
"epoch": 4.89,
"learning_rate": 4.501607717041801e-06,
"loss": 0.0228,
"step": 1520
},
{
"epoch": 4.92,
"learning_rate": 3.215434083601286e-06,
"loss": 0.0159,
"step": 1530
},
{
"epoch": 4.95,
"learning_rate": 1.929260450160772e-06,
"loss": 0.0277,
"step": 1540
},
{
"epoch": 4.98,
"learning_rate": 6.430868167202573e-07,
"loss": 0.0161,
"step": 1550
},
{
"epoch": 5.0,
"eval_accuracy": 0.9934510250569476,
"eval_loss": 0.020224373787641525,
"eval_runtime": 24.2393,
"eval_samples_per_second": 144.888,
"eval_steps_per_second": 2.269,
"step": 1555
},
{
"epoch": 5.0,
"step": 1555,
"total_flos": 0.0,
"train_loss": 0.05366519431784222,
"train_runtime": 840.8866,
"train_samples_per_second": 118.316,
"train_steps_per_second": 1.849
}
],
"max_steps": 1555,
"num_train_epochs": 5,
"total_flos": 0.0,
"trial_name": null,
"trial_params": null
}