xihaad's picture
End of training
c21ed0b verified
{
"best_metric": 0.9704,
"best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-eurosat/checkpoint-1053",
"epoch": 2.9936034115138592,
"eval_steps": 500,
"global_step": 1053,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.028429282160625444,
"grad_norm": 5.52622652053833,
"learning_rate": 4.716981132075472e-06,
"loss": 2.3451,
"step": 10
},
{
"epoch": 0.05685856432125089,
"grad_norm": 5.207578182220459,
"learning_rate": 9.433962264150944e-06,
"loss": 2.2737,
"step": 20
},
{
"epoch": 0.08528784648187633,
"grad_norm": 6.15794563293457,
"learning_rate": 1.4150943396226415e-05,
"loss": 2.1375,
"step": 30
},
{
"epoch": 0.11371712864250177,
"grad_norm": 7.397068023681641,
"learning_rate": 1.8867924528301888e-05,
"loss": 1.9859,
"step": 40
},
{
"epoch": 0.14214641080312723,
"grad_norm": 11.379995346069336,
"learning_rate": 2.358490566037736e-05,
"loss": 1.7134,
"step": 50
},
{
"epoch": 0.17057569296375266,
"grad_norm": 17.0799560546875,
"learning_rate": 2.830188679245283e-05,
"loss": 1.3856,
"step": 60
},
{
"epoch": 0.19900497512437812,
"grad_norm": 16.41306495666504,
"learning_rate": 3.30188679245283e-05,
"loss": 1.1398,
"step": 70
},
{
"epoch": 0.22743425728500355,
"grad_norm": 15.939576148986816,
"learning_rate": 3.7735849056603776e-05,
"loss": 1.0317,
"step": 80
},
{
"epoch": 0.255863539445629,
"grad_norm": 10.1390380859375,
"learning_rate": 4.245283018867925e-05,
"loss": 0.9435,
"step": 90
},
{
"epoch": 0.28429282160625446,
"grad_norm": 20.10552406311035,
"learning_rate": 4.716981132075472e-05,
"loss": 0.8801,
"step": 100
},
{
"epoch": 0.31272210376687987,
"grad_norm": 12.014751434326172,
"learning_rate": 4.978880675818374e-05,
"loss": 0.7564,
"step": 110
},
{
"epoch": 0.3411513859275053,
"grad_norm": 14.332877159118652,
"learning_rate": 4.9260823653643085e-05,
"loss": 0.6985,
"step": 120
},
{
"epoch": 0.3695806680881308,
"grad_norm": 11.214934349060059,
"learning_rate": 4.8732840549102435e-05,
"loss": 0.6713,
"step": 130
},
{
"epoch": 0.39800995024875624,
"grad_norm": 11.159377098083496,
"learning_rate": 4.820485744456177e-05,
"loss": 0.6749,
"step": 140
},
{
"epoch": 0.42643923240938164,
"grad_norm": 9.342412948608398,
"learning_rate": 4.767687434002112e-05,
"loss": 0.6158,
"step": 150
},
{
"epoch": 0.4548685145700071,
"grad_norm": 9.084425926208496,
"learning_rate": 4.7148891235480466e-05,
"loss": 0.6129,
"step": 160
},
{
"epoch": 0.48329779673063256,
"grad_norm": 11.425093650817871,
"learning_rate": 4.662090813093981e-05,
"loss": 0.6107,
"step": 170
},
{
"epoch": 0.511727078891258,
"grad_norm": 17.25963020324707,
"learning_rate": 4.609292502639916e-05,
"loss": 0.583,
"step": 180
},
{
"epoch": 0.5401563610518835,
"grad_norm": 9.029903411865234,
"learning_rate": 4.55649419218585e-05,
"loss": 0.5778,
"step": 190
},
{
"epoch": 0.5685856432125089,
"grad_norm": 11.78687858581543,
"learning_rate": 4.503695881731785e-05,
"loss": 0.6027,
"step": 200
},
{
"epoch": 0.5970149253731343,
"grad_norm": 10.835305213928223,
"learning_rate": 4.45089757127772e-05,
"loss": 0.5801,
"step": 210
},
{
"epoch": 0.6254442075337597,
"grad_norm": 9.951684951782227,
"learning_rate": 4.398099260823654e-05,
"loss": 0.5272,
"step": 220
},
{
"epoch": 0.6538734896943852,
"grad_norm": 14.171173095703125,
"learning_rate": 4.3453009503695884e-05,
"loss": 0.5197,
"step": 230
},
{
"epoch": 0.6823027718550106,
"grad_norm": 13.092684745788574,
"learning_rate": 4.292502639915523e-05,
"loss": 0.5628,
"step": 240
},
{
"epoch": 0.7107320540156361,
"grad_norm": 10.938400268554688,
"learning_rate": 4.239704329461457e-05,
"loss": 0.5621,
"step": 250
},
{
"epoch": 0.7391613361762616,
"grad_norm": 7.700541019439697,
"learning_rate": 4.186906019007392e-05,
"loss": 0.5223,
"step": 260
},
{
"epoch": 0.767590618336887,
"grad_norm": 14.169013023376465,
"learning_rate": 4.1341077085533265e-05,
"loss": 0.5656,
"step": 270
},
{
"epoch": 0.7960199004975125,
"grad_norm": 9.239885330200195,
"learning_rate": 4.081309398099261e-05,
"loss": 0.4799,
"step": 280
},
{
"epoch": 0.8244491826581379,
"grad_norm": 12.713397979736328,
"learning_rate": 4.028511087645195e-05,
"loss": 0.4996,
"step": 290
},
{
"epoch": 0.8528784648187633,
"grad_norm": 11.337251663208008,
"learning_rate": 3.97571277719113e-05,
"loss": 0.4954,
"step": 300
},
{
"epoch": 0.8813077469793887,
"grad_norm": 11.150245666503906,
"learning_rate": 3.9229144667370646e-05,
"loss": 0.4993,
"step": 310
},
{
"epoch": 0.9097370291400142,
"grad_norm": 10.354619026184082,
"learning_rate": 3.870116156282999e-05,
"loss": 0.4576,
"step": 320
},
{
"epoch": 0.9381663113006397,
"grad_norm": 9.165169715881348,
"learning_rate": 3.817317845828934e-05,
"loss": 0.4748,
"step": 330
},
{
"epoch": 0.9665955934612651,
"grad_norm": 8.658796310424805,
"learning_rate": 3.764519535374868e-05,
"loss": 0.4578,
"step": 340
},
{
"epoch": 0.9950248756218906,
"grad_norm": 11.11921215057373,
"learning_rate": 3.711721224920803e-05,
"loss": 0.502,
"step": 350
},
{
"epoch": 0.997867803837953,
"eval_accuracy": 0.9496,
"eval_loss": 0.15361838042736053,
"eval_runtime": 30.4943,
"eval_samples_per_second": 163.965,
"eval_steps_per_second": 5.148,
"step": 351
},
{
"epoch": 1.023454157782516,
"grad_norm": 10.098979949951172,
"learning_rate": 3.658922914466738e-05,
"loss": 0.4338,
"step": 360
},
{
"epoch": 1.0518834399431414,
"grad_norm": 9.755829811096191,
"learning_rate": 3.6061246040126714e-05,
"loss": 0.4701,
"step": 370
},
{
"epoch": 1.080312722103767,
"grad_norm": 7.626587867736816,
"learning_rate": 3.5533262935586064e-05,
"loss": 0.4872,
"step": 380
},
{
"epoch": 1.1087420042643923,
"grad_norm": 8.716501235961914,
"learning_rate": 3.500527983104541e-05,
"loss": 0.4495,
"step": 390
},
{
"epoch": 1.1371712864250179,
"grad_norm": 10.042838096618652,
"learning_rate": 3.447729672650475e-05,
"loss": 0.4688,
"step": 400
},
{
"epoch": 1.1656005685856432,
"grad_norm": 9.901670455932617,
"learning_rate": 3.3949313621964095e-05,
"loss": 0.4572,
"step": 410
},
{
"epoch": 1.1940298507462686,
"grad_norm": 12.942643165588379,
"learning_rate": 3.3421330517423445e-05,
"loss": 0.4174,
"step": 420
},
{
"epoch": 1.2224591329068941,
"grad_norm": 8.440229415893555,
"learning_rate": 3.289334741288279e-05,
"loss": 0.4433,
"step": 430
},
{
"epoch": 1.2508884150675195,
"grad_norm": 8.764989852905273,
"learning_rate": 3.236536430834213e-05,
"loss": 0.4542,
"step": 440
},
{
"epoch": 1.279317697228145,
"grad_norm": 11.38581657409668,
"learning_rate": 3.183738120380148e-05,
"loss": 0.4038,
"step": 450
},
{
"epoch": 1.3077469793887704,
"grad_norm": 10.445256233215332,
"learning_rate": 3.130939809926082e-05,
"loss": 0.4418,
"step": 460
},
{
"epoch": 1.336176261549396,
"grad_norm": 12.250975608825684,
"learning_rate": 3.078141499472017e-05,
"loss": 0.4717,
"step": 470
},
{
"epoch": 1.3646055437100213,
"grad_norm": 11.0380859375,
"learning_rate": 3.0253431890179517e-05,
"loss": 0.421,
"step": 480
},
{
"epoch": 1.3930348258706466,
"grad_norm": 12.513349533081055,
"learning_rate": 2.972544878563886e-05,
"loss": 0.4719,
"step": 490
},
{
"epoch": 1.4214641080312722,
"grad_norm": 11.536993980407715,
"learning_rate": 2.9197465681098207e-05,
"loss": 0.4042,
"step": 500
},
{
"epoch": 1.4498933901918978,
"grad_norm": 7.970879077911377,
"learning_rate": 2.8669482576557548e-05,
"loss": 0.415,
"step": 510
},
{
"epoch": 1.4783226723525231,
"grad_norm": 10.883822441101074,
"learning_rate": 2.8141499472016898e-05,
"loss": 0.4353,
"step": 520
},
{
"epoch": 1.5067519545131485,
"grad_norm": 14.611568450927734,
"learning_rate": 2.7613516367476245e-05,
"loss": 0.4385,
"step": 530
},
{
"epoch": 1.535181236673774,
"grad_norm": 9.424297332763672,
"learning_rate": 2.7085533262935585e-05,
"loss": 0.4712,
"step": 540
},
{
"epoch": 1.5636105188343994,
"grad_norm": 9.887263298034668,
"learning_rate": 2.6557550158394935e-05,
"loss": 0.4291,
"step": 550
},
{
"epoch": 1.5920398009950247,
"grad_norm": 8.692890167236328,
"learning_rate": 2.6029567053854276e-05,
"loss": 0.3915,
"step": 560
},
{
"epoch": 1.6204690831556503,
"grad_norm": 10.517526626586914,
"learning_rate": 2.5501583949313622e-05,
"loss": 0.4303,
"step": 570
},
{
"epoch": 1.6488983653162759,
"grad_norm": 8.96032428741455,
"learning_rate": 2.497360084477297e-05,
"loss": 0.43,
"step": 580
},
{
"epoch": 1.6773276474769012,
"grad_norm": 8.543038368225098,
"learning_rate": 2.4445617740232313e-05,
"loss": 0.4177,
"step": 590
},
{
"epoch": 1.7057569296375266,
"grad_norm": 10.253332138061523,
"learning_rate": 2.391763463569166e-05,
"loss": 0.4033,
"step": 600
},
{
"epoch": 1.7341862117981521,
"grad_norm": 9.046222686767578,
"learning_rate": 2.3389651531151003e-05,
"loss": 0.4031,
"step": 610
},
{
"epoch": 1.7626154939587777,
"grad_norm": 8.473087310791016,
"learning_rate": 2.286166842661035e-05,
"loss": 0.4012,
"step": 620
},
{
"epoch": 1.7910447761194028,
"grad_norm": 9.883254051208496,
"learning_rate": 2.2333685322069694e-05,
"loss": 0.3694,
"step": 630
},
{
"epoch": 1.8194740582800284,
"grad_norm": 10.291656494140625,
"learning_rate": 2.180570221752904e-05,
"loss": 0.3938,
"step": 640
},
{
"epoch": 1.847903340440654,
"grad_norm": 9.967813491821289,
"learning_rate": 2.1277719112988384e-05,
"loss": 0.4355,
"step": 650
},
{
"epoch": 1.8763326226012793,
"grad_norm": 11.451550483703613,
"learning_rate": 2.074973600844773e-05,
"loss": 0.4208,
"step": 660
},
{
"epoch": 1.9047619047619047,
"grad_norm": 6.351629257202148,
"learning_rate": 2.0221752903907075e-05,
"loss": 0.4091,
"step": 670
},
{
"epoch": 1.9331911869225302,
"grad_norm": 11.102736473083496,
"learning_rate": 1.9693769799366422e-05,
"loss": 0.3957,
"step": 680
},
{
"epoch": 1.9616204690831558,
"grad_norm": 7.058555603027344,
"learning_rate": 1.9165786694825765e-05,
"loss": 0.4503,
"step": 690
},
{
"epoch": 1.9900497512437811,
"grad_norm": 8.091329574584961,
"learning_rate": 1.863780359028511e-05,
"loss": 0.3782,
"step": 700
},
{
"epoch": 1.9985785358919688,
"eval_accuracy": 0.9656,
"eval_loss": 0.10786179453134537,
"eval_runtime": 30.9276,
"eval_samples_per_second": 161.668,
"eval_steps_per_second": 5.076,
"step": 703
},
{
"epoch": 2.0184790334044065,
"grad_norm": 8.83482551574707,
"learning_rate": 1.810982048574446e-05,
"loss": 0.3938,
"step": 710
},
{
"epoch": 2.046908315565032,
"grad_norm": 8.442475318908691,
"learning_rate": 1.7581837381203803e-05,
"loss": 0.379,
"step": 720
},
{
"epoch": 2.0753375977256576,
"grad_norm": 12.47827434539795,
"learning_rate": 1.7053854276663146e-05,
"loss": 0.3684,
"step": 730
},
{
"epoch": 2.1037668798862827,
"grad_norm": 8.256750106811523,
"learning_rate": 1.6525871172122493e-05,
"loss": 0.3741,
"step": 740
},
{
"epoch": 2.1321961620469083,
"grad_norm": 6.83493185043335,
"learning_rate": 1.5997888067581837e-05,
"loss": 0.361,
"step": 750
},
{
"epoch": 2.160625444207534,
"grad_norm": 7.63658332824707,
"learning_rate": 1.5469904963041184e-05,
"loss": 0.3199,
"step": 760
},
{
"epoch": 2.189054726368159,
"grad_norm": 10.230888366699219,
"learning_rate": 1.4941921858500529e-05,
"loss": 0.3495,
"step": 770
},
{
"epoch": 2.2174840085287846,
"grad_norm": 9.374876022338867,
"learning_rate": 1.4413938753959874e-05,
"loss": 0.3852,
"step": 780
},
{
"epoch": 2.24591329068941,
"grad_norm": 6.269606590270996,
"learning_rate": 1.388595564941922e-05,
"loss": 0.3398,
"step": 790
},
{
"epoch": 2.2743425728500357,
"grad_norm": 10.073060989379883,
"learning_rate": 1.3357972544878563e-05,
"loss": 0.3876,
"step": 800
},
{
"epoch": 2.302771855010661,
"grad_norm": 9.419556617736816,
"learning_rate": 1.2829989440337912e-05,
"loss": 0.3517,
"step": 810
},
{
"epoch": 2.3312011371712864,
"grad_norm": 10.461442947387695,
"learning_rate": 1.2302006335797255e-05,
"loss": 0.367,
"step": 820
},
{
"epoch": 2.359630419331912,
"grad_norm": 7.293799877166748,
"learning_rate": 1.17740232312566e-05,
"loss": 0.3402,
"step": 830
},
{
"epoch": 2.388059701492537,
"grad_norm": 6.696635723114014,
"learning_rate": 1.1246040126715946e-05,
"loss": 0.3578,
"step": 840
},
{
"epoch": 2.4164889836531627,
"grad_norm": 8.069903373718262,
"learning_rate": 1.0718057022175291e-05,
"loss": 0.3638,
"step": 850
},
{
"epoch": 2.4449182658137882,
"grad_norm": 8.903417587280273,
"learning_rate": 1.0190073917634636e-05,
"loss": 0.3618,
"step": 860
},
{
"epoch": 2.473347547974414,
"grad_norm": 10.900948524475098,
"learning_rate": 9.662090813093982e-06,
"loss": 0.3302,
"step": 870
},
{
"epoch": 2.501776830135039,
"grad_norm": 7.725928783416748,
"learning_rate": 9.134107708553327e-06,
"loss": 0.3483,
"step": 880
},
{
"epoch": 2.5302061122956645,
"grad_norm": 7.297862529754639,
"learning_rate": 8.606124604012672e-06,
"loss": 0.4057,
"step": 890
},
{
"epoch": 2.55863539445629,
"grad_norm": 9.81568717956543,
"learning_rate": 8.078141499472017e-06,
"loss": 0.3726,
"step": 900
},
{
"epoch": 2.587064676616915,
"grad_norm": 11.697639465332031,
"learning_rate": 7.5501583949313625e-06,
"loss": 0.3695,
"step": 910
},
{
"epoch": 2.6154939587775408,
"grad_norm": 8.077279090881348,
"learning_rate": 7.022175290390708e-06,
"loss": 0.3723,
"step": 920
},
{
"epoch": 2.6439232409381663,
"grad_norm": 8.867374420166016,
"learning_rate": 6.494192185850054e-06,
"loss": 0.3785,
"step": 930
},
{
"epoch": 2.672352523098792,
"grad_norm": 6.055303573608398,
"learning_rate": 5.966209081309398e-06,
"loss": 0.3597,
"step": 940
},
{
"epoch": 2.7007818052594175,
"grad_norm": 7.143568992614746,
"learning_rate": 5.438225976768744e-06,
"loss": 0.3668,
"step": 950
},
{
"epoch": 2.7292110874200426,
"grad_norm": 7.816047668457031,
"learning_rate": 4.910242872228089e-06,
"loss": 0.3296,
"step": 960
},
{
"epoch": 2.757640369580668,
"grad_norm": 8.865290641784668,
"learning_rate": 4.382259767687434e-06,
"loss": 0.3502,
"step": 970
},
{
"epoch": 2.7860696517412933,
"grad_norm": 12.268588066101074,
"learning_rate": 3.854276663146779e-06,
"loss": 0.34,
"step": 980
},
{
"epoch": 2.814498933901919,
"grad_norm": 11.221428871154785,
"learning_rate": 3.326293558606125e-06,
"loss": 0.359,
"step": 990
},
{
"epoch": 2.8429282160625444,
"grad_norm": 7.872012138366699,
"learning_rate": 2.79831045406547e-06,
"loss": 0.3483,
"step": 1000
},
{
"epoch": 2.87135749822317,
"grad_norm": 7.7068257331848145,
"learning_rate": 2.2703273495248154e-06,
"loss": 0.3468,
"step": 1010
},
{
"epoch": 2.8997867803837956,
"grad_norm": 7.7084503173828125,
"learning_rate": 1.7423442449841606e-06,
"loss": 0.3782,
"step": 1020
},
{
"epoch": 2.9282160625444207,
"grad_norm": 11.527771949768066,
"learning_rate": 1.2143611404435059e-06,
"loss": 0.3441,
"step": 1030
},
{
"epoch": 2.9566453447050463,
"grad_norm": 8.335259437561035,
"learning_rate": 6.863780359028511e-07,
"loss": 0.3625,
"step": 1040
},
{
"epoch": 2.9850746268656714,
"grad_norm": 8.47817611694336,
"learning_rate": 1.5839493136219642e-07,
"loss": 0.3119,
"step": 1050
},
{
"epoch": 2.9936034115138592,
"eval_accuracy": 0.9704,
"eval_loss": 0.09273476898670197,
"eval_runtime": 31.2742,
"eval_samples_per_second": 159.876,
"eval_steps_per_second": 5.02,
"step": 1053
},
{
"epoch": 2.9936034115138592,
"step": 1053,
"total_flos": 3.3497451642252165e+18,
"train_loss": 0.5471523041285907,
"train_runtime": 2023.4436,
"train_samples_per_second": 66.718,
"train_steps_per_second": 0.52
}
],
"logging_steps": 10,
"max_steps": 1053,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 3.3497451642252165e+18,
"train_batch_size": 32,
"trial_name": null,
"trial_params": null
}