lu5's picture
Model save
4327878
{
"best_metric": 0.8846732673267327,
"best_model_checkpoint": "swinv2-small-patch4-window8-256-finetuned-eurosat/checkpoint-1776",
"epoch": 3.0,
"eval_steps": 500,
"global_step": 1776,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.02,
"learning_rate": 5.617977528089888e-06,
"loss": 4.6512,
"step": 10
},
{
"epoch": 0.03,
"learning_rate": 1.1235955056179776e-05,
"loss": 4.6382,
"step": 20
},
{
"epoch": 0.05,
"learning_rate": 1.6853932584269665e-05,
"loss": 4.5931,
"step": 30
},
{
"epoch": 0.07,
"learning_rate": 2.2471910112359552e-05,
"loss": 4.5413,
"step": 40
},
{
"epoch": 0.08,
"learning_rate": 2.8089887640449443e-05,
"loss": 4.4003,
"step": 50
},
{
"epoch": 0.1,
"learning_rate": 3.370786516853933e-05,
"loss": 4.1375,
"step": 60
},
{
"epoch": 0.12,
"learning_rate": 3.9325842696629214e-05,
"loss": 3.7423,
"step": 70
},
{
"epoch": 0.14,
"learning_rate": 4.4943820224719104e-05,
"loss": 3.3238,
"step": 80
},
{
"epoch": 0.15,
"learning_rate": 5.0561797752808995e-05,
"loss": 2.8164,
"step": 90
},
{
"epoch": 0.17,
"learning_rate": 5.6179775280898885e-05,
"loss": 2.4727,
"step": 100
},
{
"epoch": 0.19,
"learning_rate": 6.179775280898876e-05,
"loss": 2.2967,
"step": 110
},
{
"epoch": 0.2,
"learning_rate": 6.741573033707866e-05,
"loss": 2.1826,
"step": 120
},
{
"epoch": 0.22,
"learning_rate": 7.303370786516854e-05,
"loss": 2.0388,
"step": 130
},
{
"epoch": 0.24,
"learning_rate": 7.865168539325843e-05,
"loss": 1.9666,
"step": 140
},
{
"epoch": 0.25,
"learning_rate": 8.426966292134831e-05,
"loss": 1.8424,
"step": 150
},
{
"epoch": 0.27,
"learning_rate": 8.988764044943821e-05,
"loss": 1.7991,
"step": 160
},
{
"epoch": 0.29,
"learning_rate": 9.550561797752809e-05,
"loss": 1.6829,
"step": 170
},
{
"epoch": 0.3,
"learning_rate": 9.987484355444305e-05,
"loss": 1.7742,
"step": 180
},
{
"epoch": 0.32,
"learning_rate": 9.924906132665834e-05,
"loss": 1.7502,
"step": 190
},
{
"epoch": 0.34,
"learning_rate": 9.86232790988736e-05,
"loss": 1.7302,
"step": 200
},
{
"epoch": 0.35,
"learning_rate": 9.799749687108887e-05,
"loss": 1.6316,
"step": 210
},
{
"epoch": 0.37,
"learning_rate": 9.737171464330413e-05,
"loss": 1.7153,
"step": 220
},
{
"epoch": 0.39,
"learning_rate": 9.67459324155194e-05,
"loss": 1.5892,
"step": 230
},
{
"epoch": 0.41,
"learning_rate": 9.612015018773467e-05,
"loss": 1.6424,
"step": 240
},
{
"epoch": 0.42,
"learning_rate": 9.549436795994994e-05,
"loss": 1.6096,
"step": 250
},
{
"epoch": 0.44,
"learning_rate": 9.486858573216522e-05,
"loss": 1.5543,
"step": 260
},
{
"epoch": 0.46,
"learning_rate": 9.424280350438049e-05,
"loss": 1.5115,
"step": 270
},
{
"epoch": 0.47,
"learning_rate": 9.361702127659576e-05,
"loss": 1.6442,
"step": 280
},
{
"epoch": 0.49,
"learning_rate": 9.299123904881102e-05,
"loss": 1.5456,
"step": 290
},
{
"epoch": 0.51,
"learning_rate": 9.236545682102628e-05,
"loss": 1.5322,
"step": 300
},
{
"epoch": 0.52,
"learning_rate": 9.173967459324155e-05,
"loss": 1.5012,
"step": 310
},
{
"epoch": 0.54,
"learning_rate": 9.111389236545682e-05,
"loss": 1.5726,
"step": 320
},
{
"epoch": 0.56,
"learning_rate": 9.04881101376721e-05,
"loss": 1.5606,
"step": 330
},
{
"epoch": 0.57,
"learning_rate": 8.986232790988737e-05,
"loss": 1.5696,
"step": 340
},
{
"epoch": 0.59,
"learning_rate": 8.923654568210264e-05,
"loss": 1.5081,
"step": 350
},
{
"epoch": 0.61,
"learning_rate": 8.86107634543179e-05,
"loss": 1.3746,
"step": 360
},
{
"epoch": 0.62,
"learning_rate": 8.798498122653317e-05,
"loss": 1.5299,
"step": 370
},
{
"epoch": 0.64,
"learning_rate": 8.735919899874844e-05,
"loss": 1.5458,
"step": 380
},
{
"epoch": 0.66,
"learning_rate": 8.67334167709637e-05,
"loss": 1.4804,
"step": 390
},
{
"epoch": 0.68,
"learning_rate": 8.610763454317898e-05,
"loss": 1.5249,
"step": 400
},
{
"epoch": 0.69,
"learning_rate": 8.548185231539425e-05,
"loss": 1.4581,
"step": 410
},
{
"epoch": 0.71,
"learning_rate": 8.485607008760952e-05,
"loss": 1.4566,
"step": 420
},
{
"epoch": 0.73,
"learning_rate": 8.423028785982479e-05,
"loss": 1.4358,
"step": 430
},
{
"epoch": 0.74,
"learning_rate": 8.360450563204006e-05,
"loss": 1.4832,
"step": 440
},
{
"epoch": 0.76,
"learning_rate": 8.297872340425533e-05,
"loss": 1.4909,
"step": 450
},
{
"epoch": 0.78,
"learning_rate": 8.23529411764706e-05,
"loss": 1.4253,
"step": 460
},
{
"epoch": 0.79,
"learning_rate": 8.172715894868585e-05,
"loss": 1.3965,
"step": 470
},
{
"epoch": 0.81,
"learning_rate": 8.110137672090113e-05,
"loss": 1.4092,
"step": 480
},
{
"epoch": 0.83,
"learning_rate": 8.04755944931164e-05,
"loss": 1.5548,
"step": 490
},
{
"epoch": 0.84,
"learning_rate": 7.984981226533167e-05,
"loss": 1.4722,
"step": 500
},
{
"epoch": 0.86,
"learning_rate": 7.922403003754694e-05,
"loss": 1.4116,
"step": 510
},
{
"epoch": 0.88,
"learning_rate": 7.859824780976221e-05,
"loss": 1.3771,
"step": 520
},
{
"epoch": 0.9,
"learning_rate": 7.797246558197748e-05,
"loss": 1.3767,
"step": 530
},
{
"epoch": 0.91,
"learning_rate": 7.734668335419275e-05,
"loss": 1.4398,
"step": 540
},
{
"epoch": 0.93,
"learning_rate": 7.672090112640801e-05,
"loss": 1.3905,
"step": 550
},
{
"epoch": 0.95,
"learning_rate": 7.609511889862328e-05,
"loss": 1.4094,
"step": 560
},
{
"epoch": 0.96,
"learning_rate": 7.546933667083855e-05,
"loss": 1.3176,
"step": 570
},
{
"epoch": 0.98,
"learning_rate": 7.484355444305382e-05,
"loss": 1.3472,
"step": 580
},
{
"epoch": 1.0,
"learning_rate": 7.421777221526909e-05,
"loss": 1.3521,
"step": 590
},
{
"epoch": 1.0,
"eval_accuracy": 0.7967128712871288,
"eval_loss": 0.7232595682144165,
"eval_runtime": 410.407,
"eval_samples_per_second": 61.524,
"eval_steps_per_second": 1.925,
"step": 592
},
{
"epoch": 1.01,
"learning_rate": 7.359198998748436e-05,
"loss": 1.2308,
"step": 600
},
{
"epoch": 1.03,
"learning_rate": 7.296620775969963e-05,
"loss": 1.2234,
"step": 610
},
{
"epoch": 1.05,
"learning_rate": 7.23404255319149e-05,
"loss": 1.26,
"step": 620
},
{
"epoch": 1.06,
"learning_rate": 7.171464330413017e-05,
"loss": 1.2591,
"step": 630
},
{
"epoch": 1.08,
"learning_rate": 7.108886107634543e-05,
"loss": 1.2423,
"step": 640
},
{
"epoch": 1.1,
"learning_rate": 7.04630788485607e-05,
"loss": 1.1787,
"step": 650
},
{
"epoch": 1.11,
"learning_rate": 6.983729662077597e-05,
"loss": 1.1633,
"step": 660
},
{
"epoch": 1.13,
"learning_rate": 6.921151439299124e-05,
"loss": 1.157,
"step": 670
},
{
"epoch": 1.15,
"learning_rate": 6.858573216520651e-05,
"loss": 1.2884,
"step": 680
},
{
"epoch": 1.17,
"learning_rate": 6.795994993742179e-05,
"loss": 1.2148,
"step": 690
},
{
"epoch": 1.18,
"learning_rate": 6.733416770963705e-05,
"loss": 1.1711,
"step": 700
},
{
"epoch": 1.2,
"learning_rate": 6.670838548185232e-05,
"loss": 1.2309,
"step": 710
},
{
"epoch": 1.22,
"learning_rate": 6.608260325406758e-05,
"loss": 1.1382,
"step": 720
},
{
"epoch": 1.23,
"learning_rate": 6.545682102628285e-05,
"loss": 1.1452,
"step": 730
},
{
"epoch": 1.25,
"learning_rate": 6.483103879849812e-05,
"loss": 1.2613,
"step": 740
},
{
"epoch": 1.27,
"learning_rate": 6.420525657071339e-05,
"loss": 1.1829,
"step": 750
},
{
"epoch": 1.28,
"learning_rate": 6.357947434292867e-05,
"loss": 1.0893,
"step": 760
},
{
"epoch": 1.3,
"learning_rate": 6.295369211514394e-05,
"loss": 1.1693,
"step": 770
},
{
"epoch": 1.32,
"learning_rate": 6.23279098873592e-05,
"loss": 1.2515,
"step": 780
},
{
"epoch": 1.33,
"learning_rate": 6.170212765957447e-05,
"loss": 1.2137,
"step": 790
},
{
"epoch": 1.35,
"learning_rate": 6.107634543178974e-05,
"loss": 1.1885,
"step": 800
},
{
"epoch": 1.37,
"learning_rate": 6.0450563204005004e-05,
"loss": 1.2023,
"step": 810
},
{
"epoch": 1.39,
"learning_rate": 5.982478097622027e-05,
"loss": 1.1222,
"step": 820
},
{
"epoch": 1.4,
"learning_rate": 5.9198998748435556e-05,
"loss": 1.2426,
"step": 830
},
{
"epoch": 1.42,
"learning_rate": 5.857321652065082e-05,
"loss": 1.2031,
"step": 840
},
{
"epoch": 1.44,
"learning_rate": 5.7947434292866087e-05,
"loss": 1.1798,
"step": 850
},
{
"epoch": 1.45,
"learning_rate": 5.7321652065081355e-05,
"loss": 1.1599,
"step": 860
},
{
"epoch": 1.47,
"learning_rate": 5.6695869837296624e-05,
"loss": 1.129,
"step": 870
},
{
"epoch": 1.49,
"learning_rate": 5.607008760951189e-05,
"loss": 1.155,
"step": 880
},
{
"epoch": 1.5,
"learning_rate": 5.5444305381727155e-05,
"loss": 1.1369,
"step": 890
},
{
"epoch": 1.52,
"learning_rate": 5.481852315394244e-05,
"loss": 1.1286,
"step": 900
},
{
"epoch": 1.54,
"learning_rate": 5.4192740926157707e-05,
"loss": 1.1141,
"step": 910
},
{
"epoch": 1.55,
"learning_rate": 5.356695869837297e-05,
"loss": 1.1777,
"step": 920
},
{
"epoch": 1.57,
"learning_rate": 5.294117647058824e-05,
"loss": 1.0555,
"step": 930
},
{
"epoch": 1.59,
"learning_rate": 5.2315394242803506e-05,
"loss": 1.1555,
"step": 940
},
{
"epoch": 1.6,
"learning_rate": 5.1689612015018775e-05,
"loss": 1.1587,
"step": 950
},
{
"epoch": 1.62,
"learning_rate": 5.1063829787234044e-05,
"loss": 1.115,
"step": 960
},
{
"epoch": 1.64,
"learning_rate": 5.043804755944932e-05,
"loss": 1.1189,
"step": 970
},
{
"epoch": 1.66,
"learning_rate": 4.981226533166458e-05,
"loss": 1.0221,
"step": 980
},
{
"epoch": 1.67,
"learning_rate": 4.918648310387986e-05,
"loss": 1.2178,
"step": 990
},
{
"epoch": 1.69,
"learning_rate": 4.856070087609512e-05,
"loss": 1.1566,
"step": 1000
},
{
"epoch": 1.71,
"learning_rate": 4.793491864831039e-05,
"loss": 1.0625,
"step": 1010
},
{
"epoch": 1.72,
"learning_rate": 4.730913642052566e-05,
"loss": 1.1137,
"step": 1020
},
{
"epoch": 1.74,
"learning_rate": 4.668335419274093e-05,
"loss": 1.0784,
"step": 1030
},
{
"epoch": 1.76,
"learning_rate": 4.6057571964956195e-05,
"loss": 0.9601,
"step": 1040
},
{
"epoch": 1.77,
"learning_rate": 4.5431789737171464e-05,
"loss": 1.0944,
"step": 1050
},
{
"epoch": 1.79,
"learning_rate": 4.480600750938674e-05,
"loss": 1.0572,
"step": 1060
},
{
"epoch": 1.81,
"learning_rate": 4.418022528160201e-05,
"loss": 1.1746,
"step": 1070
},
{
"epoch": 1.82,
"learning_rate": 4.355444305381727e-05,
"loss": 1.0443,
"step": 1080
},
{
"epoch": 1.84,
"learning_rate": 4.292866082603254e-05,
"loss": 0.997,
"step": 1090
},
{
"epoch": 1.86,
"learning_rate": 4.2302878598247815e-05,
"loss": 1.0583,
"step": 1100
},
{
"epoch": 1.88,
"learning_rate": 4.1677096370463084e-05,
"loss": 1.0128,
"step": 1110
},
{
"epoch": 1.89,
"learning_rate": 4.1051314142678346e-05,
"loss": 1.0192,
"step": 1120
},
{
"epoch": 1.91,
"learning_rate": 4.0425531914893614e-05,
"loss": 1.0318,
"step": 1130
},
{
"epoch": 1.93,
"learning_rate": 3.979974968710889e-05,
"loss": 1.0722,
"step": 1140
},
{
"epoch": 1.94,
"learning_rate": 3.917396745932416e-05,
"loss": 1.1282,
"step": 1150
},
{
"epoch": 1.96,
"learning_rate": 3.854818523153942e-05,
"loss": 0.964,
"step": 1160
},
{
"epoch": 1.98,
"learning_rate": 3.79224030037547e-05,
"loss": 1.045,
"step": 1170
},
{
"epoch": 1.99,
"learning_rate": 3.7296620775969966e-05,
"loss": 0.9887,
"step": 1180
},
{
"epoch": 2.0,
"eval_accuracy": 0.8518811881188119,
"eval_loss": 0.5210747718811035,
"eval_runtime": 410.9839,
"eval_samples_per_second": 61.438,
"eval_steps_per_second": 1.922,
"step": 1184
},
{
"epoch": 2.01,
"learning_rate": 3.6670838548185234e-05,
"loss": 0.8949,
"step": 1190
},
{
"epoch": 2.03,
"learning_rate": 3.6045056320400496e-05,
"loss": 0.8342,
"step": 1200
},
{
"epoch": 2.04,
"learning_rate": 3.541927409261577e-05,
"loss": 0.8466,
"step": 1210
},
{
"epoch": 2.06,
"learning_rate": 3.479349186483104e-05,
"loss": 0.9382,
"step": 1220
},
{
"epoch": 2.08,
"learning_rate": 3.416770963704631e-05,
"loss": 0.8881,
"step": 1230
},
{
"epoch": 2.09,
"learning_rate": 3.354192740926158e-05,
"loss": 0.9504,
"step": 1240
},
{
"epoch": 2.11,
"learning_rate": 3.291614518147685e-05,
"loss": 1.0155,
"step": 1250
},
{
"epoch": 2.13,
"learning_rate": 3.2290362953692116e-05,
"loss": 0.879,
"step": 1260
},
{
"epoch": 2.15,
"learning_rate": 3.1664580725907385e-05,
"loss": 0.8233,
"step": 1270
},
{
"epoch": 2.16,
"learning_rate": 3.1038798498122654e-05,
"loss": 0.8813,
"step": 1280
},
{
"epoch": 2.18,
"learning_rate": 3.0413016270337923e-05,
"loss": 0.9639,
"step": 1290
},
{
"epoch": 2.2,
"learning_rate": 2.9787234042553192e-05,
"loss": 0.8352,
"step": 1300
},
{
"epoch": 2.21,
"learning_rate": 2.9161451814768464e-05,
"loss": 0.877,
"step": 1310
},
{
"epoch": 2.23,
"learning_rate": 2.8535669586983733e-05,
"loss": 0.9327,
"step": 1320
},
{
"epoch": 2.25,
"learning_rate": 2.7909887359199e-05,
"loss": 0.8177,
"step": 1330
},
{
"epoch": 2.26,
"learning_rate": 2.7284105131414267e-05,
"loss": 0.9177,
"step": 1340
},
{
"epoch": 2.28,
"learning_rate": 2.665832290362954e-05,
"loss": 0.8632,
"step": 1350
},
{
"epoch": 2.3,
"learning_rate": 2.6032540675844808e-05,
"loss": 0.8902,
"step": 1360
},
{
"epoch": 2.31,
"learning_rate": 2.5406758448060074e-05,
"loss": 0.8219,
"step": 1370
},
{
"epoch": 2.33,
"learning_rate": 2.4780976220275346e-05,
"loss": 0.8633,
"step": 1380
},
{
"epoch": 2.35,
"learning_rate": 2.4155193992490615e-05,
"loss": 0.8205,
"step": 1390
},
{
"epoch": 2.36,
"learning_rate": 2.3529411764705884e-05,
"loss": 0.9674,
"step": 1400
},
{
"epoch": 2.38,
"learning_rate": 2.2903629536921153e-05,
"loss": 0.9004,
"step": 1410
},
{
"epoch": 2.4,
"learning_rate": 2.227784730913642e-05,
"loss": 0.8426,
"step": 1420
},
{
"epoch": 2.42,
"learning_rate": 2.165206508135169e-05,
"loss": 0.8116,
"step": 1430
},
{
"epoch": 2.43,
"learning_rate": 2.102628285356696e-05,
"loss": 0.8654,
"step": 1440
},
{
"epoch": 2.45,
"learning_rate": 2.0400500625782228e-05,
"loss": 0.7668,
"step": 1450
},
{
"epoch": 2.47,
"learning_rate": 1.9774718397997497e-05,
"loss": 0.8069,
"step": 1460
},
{
"epoch": 2.48,
"learning_rate": 1.9148936170212766e-05,
"loss": 0.8381,
"step": 1470
},
{
"epoch": 2.5,
"learning_rate": 1.8523153942428038e-05,
"loss": 0.8769,
"step": 1480
},
{
"epoch": 2.52,
"learning_rate": 1.7897371714643303e-05,
"loss": 0.8304,
"step": 1490
},
{
"epoch": 2.53,
"learning_rate": 1.7271589486858576e-05,
"loss": 0.8019,
"step": 1500
},
{
"epoch": 2.55,
"learning_rate": 1.664580725907384e-05,
"loss": 0.8039,
"step": 1510
},
{
"epoch": 2.57,
"learning_rate": 1.6020025031289113e-05,
"loss": 0.7914,
"step": 1520
},
{
"epoch": 2.58,
"learning_rate": 1.539424280350438e-05,
"loss": 0.7892,
"step": 1530
},
{
"epoch": 2.6,
"learning_rate": 1.4768460575719651e-05,
"loss": 0.7891,
"step": 1540
},
{
"epoch": 2.62,
"learning_rate": 1.414267834793492e-05,
"loss": 0.8407,
"step": 1550
},
{
"epoch": 2.64,
"learning_rate": 1.3516896120150189e-05,
"loss": 0.8267,
"step": 1560
},
{
"epoch": 2.65,
"learning_rate": 1.2891113892365458e-05,
"loss": 0.7781,
"step": 1570
},
{
"epoch": 2.67,
"learning_rate": 1.2265331664580726e-05,
"loss": 0.8289,
"step": 1580
},
{
"epoch": 2.69,
"learning_rate": 1.1639549436795997e-05,
"loss": 0.8402,
"step": 1590
},
{
"epoch": 2.7,
"learning_rate": 1.1013767209011266e-05,
"loss": 0.8149,
"step": 1600
},
{
"epoch": 2.72,
"learning_rate": 1.0387984981226535e-05,
"loss": 0.693,
"step": 1610
},
{
"epoch": 2.74,
"learning_rate": 9.762202753441804e-06,
"loss": 0.8236,
"step": 1620
},
{
"epoch": 2.75,
"learning_rate": 9.136420525657072e-06,
"loss": 0.8657,
"step": 1630
},
{
"epoch": 2.77,
"learning_rate": 8.510638297872341e-06,
"loss": 0.7369,
"step": 1640
},
{
"epoch": 2.79,
"learning_rate": 7.88485607008761e-06,
"loss": 0.728,
"step": 1650
},
{
"epoch": 2.8,
"learning_rate": 7.259073842302878e-06,
"loss": 0.8345,
"step": 1660
},
{
"epoch": 2.82,
"learning_rate": 6.633291614518149e-06,
"loss": 0.7783,
"step": 1670
},
{
"epoch": 2.84,
"learning_rate": 6.007509386733417e-06,
"loss": 0.7633,
"step": 1680
},
{
"epoch": 2.85,
"learning_rate": 5.381727158948686e-06,
"loss": 0.7573,
"step": 1690
},
{
"epoch": 2.87,
"learning_rate": 4.755944931163955e-06,
"loss": 0.7043,
"step": 1700
},
{
"epoch": 2.89,
"learning_rate": 4.130162703379224e-06,
"loss": 0.7396,
"step": 1710
},
{
"epoch": 2.91,
"learning_rate": 3.5043804755944933e-06,
"loss": 0.7728,
"step": 1720
},
{
"epoch": 2.92,
"learning_rate": 2.878598247809762e-06,
"loss": 0.7644,
"step": 1730
},
{
"epoch": 2.94,
"learning_rate": 2.252816020025031e-06,
"loss": 0.6983,
"step": 1740
},
{
"epoch": 2.96,
"learning_rate": 1.6270337922403005e-06,
"loss": 0.7628,
"step": 1750
},
{
"epoch": 2.97,
"learning_rate": 1.0012515644555696e-06,
"loss": 0.7555,
"step": 1760
},
{
"epoch": 2.99,
"learning_rate": 3.7546933667083854e-07,
"loss": 0.845,
"step": 1770
},
{
"epoch": 3.0,
"eval_accuracy": 0.8846732673267327,
"eval_loss": 0.406309574842453,
"eval_runtime": 405.6983,
"eval_samples_per_second": 62.238,
"eval_steps_per_second": 1.947,
"step": 1776
},
{
"epoch": 3.0,
"step": 1776,
"total_flos": 1.3145617489821696e+19,
"train_loss": 1.3120955138861596,
"train_runtime": 9658.6122,
"train_samples_per_second": 23.528,
"train_steps_per_second": 0.184
}
],
"logging_steps": 10,
"max_steps": 1776,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 1.3145617489821696e+19,
"trial_name": null,
"trial_params": null
}