lipo_model / checkpoint-21000 /trainer_state.json
osbm's picture
Upload with huggingface_hub
2781cc5
raw
history blame
21.4 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 50.0,
"global_step": 21000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_loss": 0.7975520491600037,
"eval_mae": 0.7266795039176941,
"eval_mse": 0.7975521087646484,
"eval_rmse": 0.8930577039718628,
"eval_runtime": 2.2756,
"eval_samples_per_second": 184.568,
"eval_steps_per_second": 23.291,
"step": 420
},
{
"epoch": 1.19,
"learning_rate": 9.761904761904762e-06,
"loss": 1.111,
"step": 500
},
{
"epoch": 2.0,
"eval_loss": 0.609792947769165,
"eval_mae": 0.6158387660980225,
"eval_mse": 0.609792947769165,
"eval_rmse": 0.7808924317359924,
"eval_runtime": 1.9303,
"eval_samples_per_second": 217.582,
"eval_steps_per_second": 27.457,
"step": 840
},
{
"epoch": 2.38,
"learning_rate": 9.523809523809525e-06,
"loss": 0.4808,
"step": 1000
},
{
"epoch": 3.0,
"eval_loss": 0.6106507182121277,
"eval_mae": 0.5975062847137451,
"eval_mse": 0.6106507778167725,
"eval_rmse": 0.7814414501190186,
"eval_runtime": 2.7602,
"eval_samples_per_second": 152.163,
"eval_steps_per_second": 19.202,
"step": 1260
},
{
"epoch": 3.57,
"learning_rate": 9.285714285714288e-06,
"loss": 0.3196,
"step": 1500
},
{
"epoch": 4.0,
"eval_loss": 0.5748845338821411,
"eval_mae": 0.584411084651947,
"eval_mse": 0.5748845934867859,
"eval_rmse": 0.7582114338874817,
"eval_runtime": 1.9617,
"eval_samples_per_second": 214.099,
"eval_steps_per_second": 27.017,
"step": 1680
},
{
"epoch": 4.76,
"learning_rate": 9.047619047619049e-06,
"loss": 0.2357,
"step": 2000
},
{
"epoch": 5.0,
"eval_loss": 0.6165366172790527,
"eval_mae": 0.5906195640563965,
"eval_mse": 0.6165366172790527,
"eval_rmse": 0.785198450088501,
"eval_runtime": 2.7689,
"eval_samples_per_second": 151.687,
"eval_steps_per_second": 19.141,
"step": 2100
},
{
"epoch": 5.95,
"learning_rate": 8.80952380952381e-06,
"loss": 0.1909,
"step": 2500
},
{
"epoch": 6.0,
"eval_loss": 0.5579051375389099,
"eval_mae": 0.5814845561981201,
"eval_mse": 0.5579050779342651,
"eval_rmse": 0.7469304203987122,
"eval_runtime": 1.2975,
"eval_samples_per_second": 323.7,
"eval_steps_per_second": 40.848,
"step": 2520
},
{
"epoch": 7.0,
"eval_loss": 0.5374110341072083,
"eval_mae": 0.5643972754478455,
"eval_mse": 0.5374110341072083,
"eval_rmse": 0.7330832481384277,
"eval_runtime": 2.8385,
"eval_samples_per_second": 147.967,
"eval_steps_per_second": 18.672,
"step": 2940
},
{
"epoch": 7.14,
"learning_rate": 8.571428571428571e-06,
"loss": 0.157,
"step": 3000
},
{
"epoch": 8.0,
"eval_loss": 0.5414829850196838,
"eval_mae": 0.5653696060180664,
"eval_mse": 0.5414829254150391,
"eval_rmse": 0.735855221748352,
"eval_runtime": 2.0614,
"eval_samples_per_second": 203.747,
"eval_steps_per_second": 25.711,
"step": 3360
},
{
"epoch": 8.33,
"learning_rate": 8.333333333333334e-06,
"loss": 0.1334,
"step": 3500
},
{
"epoch": 9.0,
"eval_loss": 0.5609115958213806,
"eval_mae": 0.5715270042419434,
"eval_mse": 0.5609115362167358,
"eval_rmse": 0.7489402890205383,
"eval_runtime": 1.9712,
"eval_samples_per_second": 213.066,
"eval_steps_per_second": 26.887,
"step": 3780
},
{
"epoch": 9.52,
"learning_rate": 8.095238095238097e-06,
"loss": 0.1199,
"step": 4000
},
{
"epoch": 10.0,
"eval_loss": 0.5141286253929138,
"eval_mae": 0.5450649261474609,
"eval_mse": 0.5141286253929138,
"eval_rmse": 0.7170276045799255,
"eval_runtime": 1.9528,
"eval_samples_per_second": 215.081,
"eval_steps_per_second": 27.141,
"step": 4200
},
{
"epoch": 10.71,
"learning_rate": 7.857142857142858e-06,
"loss": 0.1112,
"step": 4500
},
{
"epoch": 11.0,
"eval_loss": 0.5070027112960815,
"eval_mae": 0.5421043634414673,
"eval_mse": 0.5070027112960815,
"eval_rmse": 0.7120411992073059,
"eval_runtime": 1.3515,
"eval_samples_per_second": 310.772,
"eval_steps_per_second": 39.216,
"step": 4620
},
{
"epoch": 11.9,
"learning_rate": 7.61904761904762e-06,
"loss": 0.0983,
"step": 5000
},
{
"epoch": 12.0,
"eval_loss": 0.5180405378341675,
"eval_mae": 0.5500118732452393,
"eval_mse": 0.5180405378341675,
"eval_rmse": 0.7197503447532654,
"eval_runtime": 1.2448,
"eval_samples_per_second": 337.399,
"eval_steps_per_second": 42.577,
"step": 5040
},
{
"epoch": 13.0,
"eval_loss": 0.5257736444473267,
"eval_mae": 0.561431348323822,
"eval_mse": 0.5257736444473267,
"eval_rmse": 0.7251024842262268,
"eval_runtime": 1.1775,
"eval_samples_per_second": 356.675,
"eval_steps_per_second": 45.009,
"step": 5460
},
{
"epoch": 13.1,
"learning_rate": 7.380952380952382e-06,
"loss": 0.0861,
"step": 5500
},
{
"epoch": 14.0,
"eval_loss": 0.5195261836051941,
"eval_mae": 0.5508889555931091,
"eval_mse": 0.5195261836051941,
"eval_rmse": 0.7207816243171692,
"eval_runtime": 1.2061,
"eval_samples_per_second": 348.216,
"eval_steps_per_second": 43.942,
"step": 5880
},
{
"epoch": 14.29,
"learning_rate": 7.1428571428571436e-06,
"loss": 0.0833,
"step": 6000
},
{
"epoch": 15.0,
"eval_loss": 0.5224341154098511,
"eval_mae": 0.552178144454956,
"eval_mse": 0.5224341154098511,
"eval_rmse": 0.7227960228919983,
"eval_runtime": 1.1671,
"eval_samples_per_second": 359.858,
"eval_steps_per_second": 45.411,
"step": 6300
},
{
"epoch": 15.48,
"learning_rate": 6.9047619047619055e-06,
"loss": 0.0741,
"step": 6500
},
{
"epoch": 16.0,
"eval_loss": 0.5121651291847229,
"eval_mae": 0.5470804572105408,
"eval_mse": 0.5121651291847229,
"eval_rmse": 0.715657114982605,
"eval_runtime": 1.1174,
"eval_samples_per_second": 375.865,
"eval_steps_per_second": 47.431,
"step": 6720
},
{
"epoch": 16.67,
"learning_rate": 6.666666666666667e-06,
"loss": 0.0744,
"step": 7000
},
{
"epoch": 17.0,
"eval_loss": 0.4920870065689087,
"eval_mae": 0.5328148007392883,
"eval_mse": 0.4920870065689087,
"eval_rmse": 0.7014891505241394,
"eval_runtime": 1.1494,
"eval_samples_per_second": 365.42,
"eval_steps_per_second": 46.112,
"step": 7140
},
{
"epoch": 17.86,
"learning_rate": 6.4285714285714295e-06,
"loss": 0.0649,
"step": 7500
},
{
"epoch": 18.0,
"eval_loss": 0.5028687119483948,
"eval_mae": 0.5357252359390259,
"eval_mse": 0.5028687119483948,
"eval_rmse": 0.7091323733329773,
"eval_runtime": 1.3327,
"eval_samples_per_second": 315.156,
"eval_steps_per_second": 39.77,
"step": 7560
},
{
"epoch": 19.0,
"eval_loss": 0.526236891746521,
"eval_mae": 0.5603699684143066,
"eval_mse": 0.526236891746521,
"eval_rmse": 0.7254218459129333,
"eval_runtime": 1.2696,
"eval_samples_per_second": 330.822,
"eval_steps_per_second": 41.747,
"step": 7980
},
{
"epoch": 19.05,
"learning_rate": 6.1904761904761914e-06,
"loss": 0.0639,
"step": 8000
},
{
"epoch": 20.0,
"eval_loss": 0.5124732851982117,
"eval_mae": 0.5440532565116882,
"eval_mse": 0.5124732851982117,
"eval_rmse": 0.7158724069595337,
"eval_runtime": 1.139,
"eval_samples_per_second": 368.742,
"eval_steps_per_second": 46.532,
"step": 8400
},
{
"epoch": 20.24,
"learning_rate": 5.9523809523809525e-06,
"loss": 0.0609,
"step": 8500
},
{
"epoch": 21.0,
"eval_loss": 0.4891616106033325,
"eval_mae": 0.5339100956916809,
"eval_mse": 0.48916158080101013,
"eval_rmse": 0.6994009017944336,
"eval_runtime": 1.0834,
"eval_samples_per_second": 387.673,
"eval_steps_per_second": 48.921,
"step": 8820
},
{
"epoch": 21.43,
"learning_rate": 5.7142857142857145e-06,
"loss": 0.0547,
"step": 9000
},
{
"epoch": 22.0,
"eval_loss": 0.4973764717578888,
"eval_mae": 0.5337387919425964,
"eval_mse": 0.4973764419555664,
"eval_rmse": 0.7052491903305054,
"eval_runtime": 1.1768,
"eval_samples_per_second": 356.886,
"eval_steps_per_second": 45.036,
"step": 9240
},
{
"epoch": 22.62,
"learning_rate": 5.476190476190477e-06,
"loss": 0.0528,
"step": 9500
},
{
"epoch": 23.0,
"eval_loss": 0.4779137670993805,
"eval_mae": 0.5240978598594666,
"eval_mse": 0.47791385650634766,
"eval_rmse": 0.6913131475448608,
"eval_runtime": 1.1884,
"eval_samples_per_second": 353.429,
"eval_steps_per_second": 44.599,
"step": 9660
},
{
"epoch": 23.81,
"learning_rate": 5.2380952380952384e-06,
"loss": 0.0514,
"step": 10000
},
{
"epoch": 24.0,
"eval_loss": 0.483572781085968,
"eval_mae": 0.5290801525115967,
"eval_mse": 0.48357275128364563,
"eval_rmse": 0.695393979549408,
"eval_runtime": 1.1926,
"eval_samples_per_second": 352.184,
"eval_steps_per_second": 44.442,
"step": 10080
},
{
"epoch": 25.0,
"learning_rate": 5e-06,
"loss": 0.0516,
"step": 10500
},
{
"epoch": 25.0,
"eval_loss": 0.4979321360588074,
"eval_mae": 0.5313663482666016,
"eval_mse": 0.49793222546577454,
"eval_rmse": 0.7056431174278259,
"eval_runtime": 1.1676,
"eval_samples_per_second": 359.71,
"eval_steps_per_second": 45.392,
"step": 10500
},
{
"epoch": 26.0,
"eval_loss": 0.48045653104782104,
"eval_mae": 0.530644416809082,
"eval_mse": 0.48045653104782104,
"eval_rmse": 0.693149745464325,
"eval_runtime": 1.1924,
"eval_samples_per_second": 352.222,
"eval_steps_per_second": 44.447,
"step": 10920
},
{
"epoch": 26.19,
"learning_rate": 4.761904761904762e-06,
"loss": 0.0463,
"step": 11000
},
{
"epoch": 27.0,
"eval_loss": 0.48534154891967773,
"eval_mae": 0.5263835191726685,
"eval_mse": 0.48534154891967773,
"eval_rmse": 0.696664571762085,
"eval_runtime": 1.1385,
"eval_samples_per_second": 368.9,
"eval_steps_per_second": 46.552,
"step": 11340
},
{
"epoch": 27.38,
"learning_rate": 4.523809523809524e-06,
"loss": 0.0451,
"step": 11500
},
{
"epoch": 28.0,
"eval_loss": 0.4952319264411926,
"eval_mae": 0.5382465720176697,
"eval_mse": 0.4952319264411926,
"eval_rmse": 0.7037271857261658,
"eval_runtime": 1.1142,
"eval_samples_per_second": 376.951,
"eval_steps_per_second": 47.568,
"step": 11760
},
{
"epoch": 28.57,
"learning_rate": 4.2857142857142855e-06,
"loss": 0.0428,
"step": 12000
},
{
"epoch": 29.0,
"eval_loss": 0.48455557227134705,
"eval_mae": 0.529035210609436,
"eval_mse": 0.48455557227134705,
"eval_rmse": 0.6961002349853516,
"eval_runtime": 1.1847,
"eval_samples_per_second": 354.512,
"eval_steps_per_second": 44.736,
"step": 12180
},
{
"epoch": 29.76,
"learning_rate": 4.047619047619048e-06,
"loss": 0.0417,
"step": 12500
},
{
"epoch": 30.0,
"eval_loss": 0.4880525767803192,
"eval_mae": 0.5353544354438782,
"eval_mse": 0.4880525767803192,
"eval_rmse": 0.6986076235771179,
"eval_runtime": 1.2654,
"eval_samples_per_second": 331.922,
"eval_steps_per_second": 41.885,
"step": 12600
},
{
"epoch": 30.95,
"learning_rate": 3.80952380952381e-06,
"loss": 0.0404,
"step": 13000
},
{
"epoch": 31.0,
"eval_loss": 0.48999953269958496,
"eval_mae": 0.5326849222183228,
"eval_mse": 0.4899995028972626,
"eval_rmse": 0.6999996304512024,
"eval_runtime": 1.2442,
"eval_samples_per_second": 337.554,
"eval_steps_per_second": 42.596,
"step": 13020
},
{
"epoch": 32.0,
"eval_loss": 0.5017576813697815,
"eval_mae": 0.5423673987388611,
"eval_mse": 0.5017576813697815,
"eval_rmse": 0.7083485722541809,
"eval_runtime": 1.1196,
"eval_samples_per_second": 375.135,
"eval_steps_per_second": 47.339,
"step": 13440
},
{
"epoch": 32.14,
"learning_rate": 3.5714285714285718e-06,
"loss": 0.0391,
"step": 13500
},
{
"epoch": 33.0,
"eval_loss": 0.48715752363204956,
"eval_mae": 0.5341249108314514,
"eval_mse": 0.48715752363204956,
"eval_rmse": 0.6979666948318481,
"eval_runtime": 1.1228,
"eval_samples_per_second": 374.062,
"eval_steps_per_second": 47.203,
"step": 13860
},
{
"epoch": 33.33,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.0373,
"step": 14000
},
{
"epoch": 34.0,
"eval_loss": 0.479116827249527,
"eval_mae": 0.5321008563041687,
"eval_mse": 0.479116827249527,
"eval_rmse": 0.6921826601028442,
"eval_runtime": 1.206,
"eval_samples_per_second": 348.254,
"eval_steps_per_second": 43.946,
"step": 14280
},
{
"epoch": 34.52,
"learning_rate": 3.0952380952380957e-06,
"loss": 0.0359,
"step": 14500
},
{
"epoch": 35.0,
"eval_loss": 0.48590707778930664,
"eval_mae": 0.5346122980117798,
"eval_mse": 0.48590704798698425,
"eval_rmse": 0.6970703601837158,
"eval_runtime": 0.8892,
"eval_samples_per_second": 472.339,
"eval_steps_per_second": 59.605,
"step": 14700
},
{
"epoch": 35.71,
"learning_rate": 2.8571428571428573e-06,
"loss": 0.0334,
"step": 15000
},
{
"epoch": 36.0,
"eval_loss": 0.4766678810119629,
"eval_mae": 0.5263535380363464,
"eval_mse": 0.4766678810119629,
"eval_rmse": 0.690411388874054,
"eval_runtime": 1.286,
"eval_samples_per_second": 326.603,
"eval_steps_per_second": 41.214,
"step": 15120
},
{
"epoch": 36.9,
"learning_rate": 2.6190476190476192e-06,
"loss": 0.0341,
"step": 15500
},
{
"epoch": 37.0,
"eval_loss": 0.4740683436393738,
"eval_mae": 0.5217899680137634,
"eval_mse": 0.4740683138370514,
"eval_rmse": 0.6885262131690979,
"eval_runtime": 1.1893,
"eval_samples_per_second": 353.162,
"eval_steps_per_second": 44.566,
"step": 15540
},
{
"epoch": 38.0,
"eval_loss": 0.478584349155426,
"eval_mae": 0.527285635471344,
"eval_mse": 0.478584349155426,
"eval_rmse": 0.6917979121208191,
"eval_runtime": 1.1477,
"eval_samples_per_second": 365.939,
"eval_steps_per_second": 46.178,
"step": 15960
},
{
"epoch": 38.1,
"learning_rate": 2.380952380952381e-06,
"loss": 0.0325,
"step": 16000
},
{
"epoch": 39.0,
"eval_loss": 0.46834418177604675,
"eval_mae": 0.5196450352668762,
"eval_mse": 0.46834418177604675,
"eval_rmse": 0.6843567490577698,
"eval_runtime": 1.1282,
"eval_samples_per_second": 372.276,
"eval_steps_per_second": 46.978,
"step": 16380
},
{
"epoch": 39.29,
"learning_rate": 2.1428571428571427e-06,
"loss": 0.0331,
"step": 16500
},
{
"epoch": 40.0,
"eval_loss": 0.46904683113098145,
"eval_mae": 0.522687554359436,
"eval_mse": 0.46904683113098145,
"eval_rmse": 0.6848699450492859,
"eval_runtime": 1.1578,
"eval_samples_per_second": 362.764,
"eval_steps_per_second": 45.777,
"step": 16800
},
{
"epoch": 40.48,
"learning_rate": 1.904761904761905e-06,
"loss": 0.0315,
"step": 17000
},
{
"epoch": 41.0,
"eval_loss": 0.46947211027145386,
"eval_mae": 0.5197708010673523,
"eval_mse": 0.46947211027145386,
"eval_rmse": 0.6851803660392761,
"eval_runtime": 1.1711,
"eval_samples_per_second": 358.64,
"eval_steps_per_second": 45.257,
"step": 17220
},
{
"epoch": 41.67,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.0299,
"step": 17500
},
{
"epoch": 42.0,
"eval_loss": 0.46514230966567993,
"eval_mae": 0.518059253692627,
"eval_mse": 0.46514227986335754,
"eval_rmse": 0.6820133924484253,
"eval_runtime": 1.1482,
"eval_samples_per_second": 365.797,
"eval_steps_per_second": 46.16,
"step": 17640
},
{
"epoch": 42.86,
"learning_rate": 1.4285714285714286e-06,
"loss": 0.0314,
"step": 18000
},
{
"epoch": 43.0,
"eval_loss": 0.4640100300312042,
"eval_mae": 0.51801997423172,
"eval_mse": 0.46400997042655945,
"eval_rmse": 0.6811828017234802,
"eval_runtime": 1.1908,
"eval_samples_per_second": 352.696,
"eval_steps_per_second": 44.507,
"step": 18060
},
{
"epoch": 44.0,
"eval_loss": 0.4704785943031311,
"eval_mae": 0.5230998992919922,
"eval_mse": 0.4704785943031311,
"eval_rmse": 0.6859143972396851,
"eval_runtime": 1.2463,
"eval_samples_per_second": 337.011,
"eval_steps_per_second": 42.528,
"step": 18480
},
{
"epoch": 44.05,
"learning_rate": 1.1904761904761906e-06,
"loss": 0.0283,
"step": 18500
},
{
"epoch": 45.0,
"eval_loss": 0.4678580164909363,
"eval_mae": 0.521861732006073,
"eval_mse": 0.46785807609558105,
"eval_rmse": 0.6840015053749084,
"eval_runtime": 1.2433,
"eval_samples_per_second": 337.812,
"eval_steps_per_second": 42.629,
"step": 18900
},
{
"epoch": 45.24,
"learning_rate": 9.523809523809525e-07,
"loss": 0.0284,
"step": 19000
},
{
"epoch": 46.0,
"eval_loss": 0.4686986207962036,
"eval_mae": 0.5229542851448059,
"eval_mse": 0.4686986207962036,
"eval_rmse": 0.6846156716346741,
"eval_runtime": 1.1892,
"eval_samples_per_second": 353.193,
"eval_steps_per_second": 44.57,
"step": 19320
},
{
"epoch": 46.43,
"learning_rate": 7.142857142857143e-07,
"loss": 0.0273,
"step": 19500
},
{
"epoch": 47.0,
"eval_loss": 0.46382173895835876,
"eval_mae": 0.5197833776473999,
"eval_mse": 0.4638217091560364,
"eval_rmse": 0.6810445785522461,
"eval_runtime": 1.2872,
"eval_samples_per_second": 326.279,
"eval_steps_per_second": 41.173,
"step": 19740
},
{
"epoch": 47.62,
"learning_rate": 4.7619047619047623e-07,
"loss": 0.0283,
"step": 20000
},
{
"epoch": 48.0,
"eval_loss": 0.4631454348564148,
"eval_mae": 0.5184262990951538,
"eval_mse": 0.4631454348564148,
"eval_rmse": 0.6805478930473328,
"eval_runtime": 1.1721,
"eval_samples_per_second": 358.345,
"eval_steps_per_second": 45.22,
"step": 20160
},
{
"epoch": 48.81,
"learning_rate": 2.3809523809523811e-07,
"loss": 0.0268,
"step": 20500
},
{
"epoch": 49.0,
"eval_loss": 0.4649342894554138,
"eval_mae": 0.5196540355682373,
"eval_mse": 0.4649342894554138,
"eval_rmse": 0.6818609237670898,
"eval_runtime": 1.1402,
"eval_samples_per_second": 368.356,
"eval_steps_per_second": 46.483,
"step": 20580
},
{
"epoch": 50.0,
"learning_rate": 0.0,
"loss": 0.0269,
"step": 21000
},
{
"epoch": 50.0,
"eval_loss": 0.46444272994995117,
"eval_mae": 0.5188572406768799,
"eval_mse": 0.46444275975227356,
"eval_rmse": 0.6815003752708435,
"eval_runtime": 1.0966,
"eval_samples_per_second": 382.987,
"eval_steps_per_second": 48.329,
"step": 21000
}
],
"max_steps": 21000,
"num_train_epochs": 50,
"total_flos": 1.1126765647872e+16,
"trial_name": null,
"trial_params": null
}