napsternxg's picture
End of training
5917a24
{
"best_metric": 0.9796159267425537,
"best_model_checkpoint": "nyt-ingredient-tagger-jina-embeddings-v2-small-en/checkpoint-25000",
"epoch": 5.0,
"eval_steps": 1000,
"global_step": 25255,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.1,
"learning_rate": 4.9010097010492975e-05,
"loss": 1.3131,
"step": 500
},
{
"epoch": 0.2,
"learning_rate": 4.802019402098595e-05,
"loss": 1.1585,
"step": 1000
},
{
"epoch": 0.2,
"eval_COMMENT": {
"f1": 0.454561770864493,
"number": 6836,
"precision": 0.38455309241826097,
"recall": 0.5557343475716794
},
"eval_NAME": {
"f1": 0.7002083333333333,
"number": 8859,
"precision": 0.6500338458563002,
"recall": 0.7587763855965685
},
"eval_QTY": {
"f1": 0.9200849140587551,
"number": 7095,
"precision": 0.8947789025039957,
"recall": 0.9468639887244539
},
"eval_RANGE_END": {
"f1": 0.0,
"number": 74,
"precision": 0.0,
"recall": 0.0
},
"eval_UNIT": {
"f1": 0.912968864917872,
"number": 5723,
"precision": 0.8575376112987412,
"recall": 0.9760615062030403
},
"eval_loss": 1.1246980428695679,
"eval_overall_accuracy": 0.7220226951438505,
"eval_overall_f1": 0.7265753075808941,
"eval_overall_precision": 0.6665790549617429,
"eval_overall_recall": 0.7984398502815965,
"eval_runtime": 10.7443,
"eval_samples_per_second": 791.676,
"eval_steps_per_second": 24.757,
"step": 1000
},
{
"epoch": 0.3,
"learning_rate": 4.703029103147892e-05,
"loss": 1.1236,
"step": 1500
},
{
"epoch": 0.4,
"learning_rate": 4.6040388041971886e-05,
"loss": 1.1018,
"step": 2000
},
{
"epoch": 0.4,
"eval_COMMENT": {
"f1": 0.48012443829934326,
"number": 6836,
"precision": 0.3960273712222011,
"recall": 0.6095669982445875
},
"eval_NAME": {
"f1": 0.733016860369946,
"number": 8859,
"precision": 0.7094423320659062,
"recall": 0.7582119878090078
},
"eval_QTY": {
"f1": 0.9238356164383562,
"number": 7095,
"precision": 0.8986009327115256,
"recall": 0.9505285412262157
},
"eval_RANGE_END": {
"f1": 0.047619047619047616,
"number": 74,
"precision": 0.2,
"recall": 0.02702702702702703
},
"eval_UNIT": {
"f1": 0.9199999999999999,
"number": 5723,
"precision": 0.8794009877329935,
"recall": 0.9645290931329722
},
"eval_loss": 1.0676864385604858,
"eval_overall_accuracy": 0.7414702173996103,
"eval_overall_f1": 0.7423559781301609,
"eval_overall_precision": 0.6852761825824404,
"eval_overall_recall": 0.8098086542834155,
"eval_runtime": 8.0746,
"eval_samples_per_second": 1053.43,
"eval_steps_per_second": 32.943,
"step": 2000
},
{
"epoch": 0.49,
"learning_rate": 4.505048505246486e-05,
"loss": 1.0885,
"step": 2500
},
{
"epoch": 0.59,
"learning_rate": 4.406058206295783e-05,
"loss": 1.0676,
"step": 3000
},
{
"epoch": 0.59,
"eval_COMMENT": {
"f1": 0.4957528957528958,
"number": 6836,
"precision": 0.41734173417341736,
"recall": 0.6104447045055588
},
"eval_NAME": {
"f1": 0.7383297644539616,
"number": 8859,
"precision": 0.7021688219122288,
"recall": 0.7784174286036799
},
"eval_QTY": {
"f1": 0.9259863945578231,
"number": 7095,
"precision": 0.8949375410913872,
"recall": 0.9592670894996477
},
"eval_RANGE_END": {
"f1": 0.3047619047619048,
"number": 74,
"precision": 0.5161290322580645,
"recall": 0.21621621621621623
},
"eval_UNIT": {
"f1": 0.9193198866477745,
"number": 5723,
"precision": 0.8788844621513944,
"recall": 0.9636554254761489
},
"eval_loss": 1.0472389459609985,
"eval_overall_accuracy": 0.7540595269934666,
"eval_overall_f1": 0.7511794345133026,
"eval_overall_precision": 0.6939017520974771,
"eval_overall_recall": 0.8187637737433099,
"eval_runtime": 7.2751,
"eval_samples_per_second": 1169.193,
"eval_steps_per_second": 36.563,
"step": 3000
},
{
"epoch": 0.69,
"learning_rate": 4.3070679073450804e-05,
"loss": 1.0667,
"step": 3500
},
{
"epoch": 0.79,
"learning_rate": 4.208077608394378e-05,
"loss": 1.0613,
"step": 4000
},
{
"epoch": 0.79,
"eval_COMMENT": {
"f1": 0.5095226317091268,
"number": 6836,
"precision": 0.4413024850042845,
"recall": 0.6026916325336454
},
"eval_NAME": {
"f1": 0.7497392545424604,
"number": 8859,
"precision": 0.7297499465697799,
"recall": 0.7708544982503669
},
"eval_QTY": {
"f1": 0.9260893546607832,
"number": 7095,
"precision": 0.9064651100013497,
"recall": 0.9465821000704722
},
"eval_RANGE_END": {
"f1": 0.23214285714285715,
"number": 74,
"precision": 0.34210526315789475,
"recall": 0.17567567567567569
},
"eval_UNIT": {
"f1": 0.9216135881104034,
"number": 5723,
"precision": 0.8965631196298744,
"recall": 0.9481041411846933
},
"eval_loss": 1.045949101448059,
"eval_overall_accuracy": 0.7502196920490581,
"eval_overall_f1": 0.7602500822639027,
"eval_overall_precision": 0.717671543503246,
"eval_overall_recall": 0.8081995312554657,
"eval_runtime": 9.0484,
"eval_samples_per_second": 940.052,
"eval_steps_per_second": 29.397,
"step": 4000
},
{
"epoch": 0.89,
"learning_rate": 4.109087309443675e-05,
"loss": 1.0532,
"step": 4500
},
{
"epoch": 0.99,
"learning_rate": 4.010097010492972e-05,
"loss": 1.045,
"step": 5000
},
{
"epoch": 0.99,
"eval_COMMENT": {
"f1": 0.5115396388123661,
"number": 6836,
"precision": 0.43983577218654596,
"recall": 0.6111761263897015
},
"eval_NAME": {
"f1": 0.7500674909562118,
"number": 8859,
"precision": 0.7188987787207618,
"recall": 0.7840614064792866
},
"eval_QTY": {
"f1": 0.9248702917593155,
"number": 7095,
"precision": 0.886005680351149,
"recall": 0.9673009161381254
},
"eval_RANGE_END": {
"f1": 0.27868852459016397,
"number": 74,
"precision": 0.3541666666666667,
"recall": 0.22972972972972974
},
"eval_UNIT": {
"f1": 0.9236628580890875,
"number": 5723,
"precision": 0.8777340676632572,
"recall": 0.974663637952123
},
"eval_loss": 1.029239296913147,
"eval_overall_accuracy": 0.7610323615940091,
"eval_overall_f1": 0.7619755400100165,
"eval_overall_precision": 0.7079555688982287,
"eval_overall_recall": 0.8249204183719873,
"eval_runtime": 10.4805,
"eval_samples_per_second": 811.605,
"eval_steps_per_second": 25.381,
"step": 5000
},
{
"epoch": 1.09,
"learning_rate": 3.911106711542269e-05,
"loss": 1.0391,
"step": 5500
},
{
"epoch": 1.19,
"learning_rate": 3.812116412591566e-05,
"loss": 1.0334,
"step": 6000
},
{
"epoch": 1.19,
"eval_COMMENT": {
"f1": 0.5539883268482491,
"number": 6836,
"precision": 0.47399084477736164,
"recall": 0.6664716208308953
},
"eval_NAME": {
"f1": 0.7479798253701395,
"number": 8859,
"precision": 0.7198329853862213,
"recall": 0.7784174286036799
},
"eval_QTY": {
"f1": 0.92860166502046,
"number": 7095,
"precision": 0.9296510806611104,
"recall": 0.927554615926709
},
"eval_RANGE_END": {
"f1": 0.3283582089552239,
"number": 74,
"precision": 0.36666666666666664,
"recall": 0.2972972972972973
},
"eval_UNIT": {
"f1": 0.9286739258884106,
"number": 5723,
"precision": 0.8982691051600261,
"recall": 0.9612091560370435
},
"eval_loss": 1.034406065940857,
"eval_overall_accuracy": 0.759465861765942,
"eval_overall_f1": 0.7717964680056355,
"eval_overall_precision": 0.7258049607148359,
"eval_overall_recall": 0.8240109140518418,
"eval_runtime": 8.1039,
"eval_samples_per_second": 1049.615,
"eval_steps_per_second": 32.824,
"step": 6000
},
{
"epoch": 1.29,
"learning_rate": 3.7131261136408633e-05,
"loss": 1.026,
"step": 6500
},
{
"epoch": 1.39,
"learning_rate": 3.6141358146901606e-05,
"loss": 1.0187,
"step": 7000
},
{
"epoch": 1.39,
"eval_COMMENT": {
"f1": 0.5136793031529874,
"number": 6836,
"precision": 0.4423198816818086,
"recall": 0.6124926857811586
},
"eval_NAME": {
"f1": 0.751153062318996,
"number": 8859,
"precision": 0.7155410238070911,
"recall": 0.7904955412574782
},
"eval_QTY": {
"f1": 0.9245735607675907,
"number": 7095,
"precision": 0.8767850372804247,
"recall": 0.9778717406624383
},
"eval_RANGE_END": {
"f1": 0.23853211009174313,
"number": 74,
"precision": 0.37142857142857144,
"recall": 0.17567567567567569
},
"eval_UNIT": {
"f1": 0.927461139896373,
"number": 5723,
"precision": 0.8888354957552459,
"recall": 0.9695963655425476
},
"eval_loss": 1.021018385887146,
"eval_overall_accuracy": 0.7651014404156956,
"eval_overall_f1": 0.7638116425658138,
"eval_overall_precision": 0.7083482836981222,
"eval_overall_recall": 0.8286983593941302,
"eval_runtime": 7.3061,
"eval_samples_per_second": 1164.235,
"eval_steps_per_second": 36.408,
"step": 7000
},
{
"epoch": 1.48,
"learning_rate": 3.515145515739457e-05,
"loss": 1.0186,
"step": 7500
},
{
"epoch": 1.58,
"learning_rate": 3.416155216788755e-05,
"loss": 1.0319,
"step": 8000
},
{
"epoch": 1.58,
"eval_COMMENT": {
"f1": 0.5439773496645535,
"number": 6836,
"precision": 0.46955690149824675,
"recall": 0.6464306612053833
},
"eval_NAME": {
"f1": 0.7617639003012875,
"number": 8859,
"precision": 0.7399957428693061,
"recall": 0.7848515633818716
},
"eval_QTY": {
"f1": 0.9297781396295542,
"number": 7095,
"precision": 0.8963893249607535,
"recall": 0.9657505285412262
},
"eval_RANGE_END": {
"f1": 0.368,
"number": 74,
"precision": 0.45098039215686275,
"recall": 0.3108108108108108
},
"eval_UNIT": {
"f1": 0.9286800743620078,
"number": 5723,
"precision": 0.8991981672394044,
"recall": 0.9601607548488555
},
"eval_loss": 1.0136394500732422,
"eval_overall_accuracy": 0.7699919764642953,
"eval_overall_f1": 0.7758823529411765,
"eval_overall_precision": 0.727991905068531,
"eval_overall_recall": 0.8305173680344212,
"eval_runtime": 9.7068,
"eval_samples_per_second": 876.293,
"eval_steps_per_second": 27.403,
"step": 8000
},
{
"epoch": 1.68,
"learning_rate": 3.317164917838052e-05,
"loss": 1.0243,
"step": 8500
},
{
"epoch": 1.78,
"learning_rate": 3.218174618887349e-05,
"loss": 1.0154,
"step": 9000
},
{
"epoch": 1.78,
"eval_COMMENT": {
"f1": 0.5554884675763797,
"number": 6836,
"precision": 0.47295907875796833,
"recall": 0.6729081334113517
},
"eval_NAME": {
"f1": 0.7644151565074134,
"number": 8859,
"precision": 0.7443054218800128,
"recall": 0.7856417202844564
},
"eval_QTY": {
"f1": 0.9317274604267033,
"number": 7095,
"precision": 0.9104236718224613,
"recall": 0.9540521494009866
},
"eval_RANGE_END": {
"f1": 0.3795620437956204,
"number": 74,
"precision": 0.4126984126984127,
"recall": 0.35135135135135137
},
"eval_UNIT": {
"f1": 0.9292205085452273,
"number": 5723,
"precision": 0.8885522959183674,
"recall": 0.9737899702952997
},
"eval_loss": 1.0071077346801758,
"eval_overall_accuracy": 0.7732204944026286,
"eval_overall_f1": 0.7789823224924309,
"eval_overall_precision": 0.7284683532742716,
"eval_overall_recall": 0.8370238220170008,
"eval_runtime": 10.018,
"eval_samples_per_second": 849.074,
"eval_steps_per_second": 26.552,
"step": 9000
},
{
"epoch": 1.88,
"learning_rate": 3.119184319936647e-05,
"loss": 1.0185,
"step": 9500
},
{
"epoch": 1.98,
"learning_rate": 3.0201940209859435e-05,
"loss": 1.011,
"step": 10000
},
{
"epoch": 1.98,
"eval_COMMENT": {
"f1": 0.5534477566997892,
"number": 6836,
"precision": 0.4703654417033473,
"recall": 0.6721767115272089
},
"eval_NAME": {
"f1": 0.7638995503892971,
"number": 8859,
"precision": 0.742723104808615,
"recall": 0.7863189976295293
},
"eval_QTY": {
"f1": 0.9353837666367836,
"number": 7095,
"precision": 0.915743991358358,
"recall": 0.9558844256518675
},
"eval_RANGE_END": {
"f1": 0.3384615384615385,
"number": 74,
"precision": 0.39285714285714285,
"recall": 0.2972972972972973
},
"eval_UNIT": {
"f1": 0.9314305016044586,
"number": 5723,
"precision": 0.901291060630822,
"recall": 0.9636554254761489
},
"eval_loss": 1.0126872062683105,
"eval_overall_accuracy": 0.7711955068200054,
"eval_overall_f1": 0.7789157805466762,
"eval_overall_precision": 0.7296281585138562,
"eval_overall_recall": 0.8353447371182705,
"eval_runtime": 7.6383,
"eval_samples_per_second": 1113.597,
"eval_steps_per_second": 34.824,
"step": 10000
},
{
"epoch": 2.08,
"learning_rate": 2.9212037220352405e-05,
"loss": 1.0062,
"step": 10500
},
{
"epoch": 2.18,
"learning_rate": 2.8222134230845377e-05,
"loss": 0.9958,
"step": 11000
},
{
"epoch": 2.18,
"eval_COMMENT": {
"f1": 0.5524928818077179,
"number": 6836,
"precision": 0.4715127701375246,
"recall": 0.6670567583382094
},
"eval_NAME": {
"f1": 0.7657111356119073,
"number": 8859,
"precision": 0.7483029845921776,
"recall": 0.7839485269217744
},
"eval_QTY": {
"f1": 0.9343237989447376,
"number": 7095,
"precision": 0.9206457791763579,
"recall": 0.9484143763213531
},
"eval_RANGE_END": {
"f1": 0.36363636363636365,
"number": 74,
"precision": 0.35,
"recall": 0.3783783783783784
},
"eval_UNIT": {
"f1": 0.9317363263254879,
"number": 5723,
"precision": 0.8947876447876448,
"recall": 0.9718679014502883
},
"eval_loss": 1.0023564100265503,
"eval_overall_accuracy": 0.7762770794330035,
"eval_overall_f1": 0.7792751537354442,
"eval_overall_precision": 0.7317627545535522,
"eval_overall_recall": 0.8333858047364187,
"eval_runtime": 8.2445,
"eval_samples_per_second": 1031.712,
"eval_steps_per_second": 32.264,
"step": 11000
},
{
"epoch": 2.28,
"learning_rate": 2.7232231241338353e-05,
"loss": 0.9969,
"step": 11500
},
{
"epoch": 2.38,
"learning_rate": 2.6242328251831323e-05,
"loss": 1.0042,
"step": 12000
},
{
"epoch": 2.38,
"eval_COMMENT": {
"f1": 0.5542747950440732,
"number": 6836,
"precision": 0.4789602641951635,
"recall": 0.657694558221182
},
"eval_NAME": {
"f1": 0.7601839329185826,
"number": 8859,
"precision": 0.7298981923955953,
"recall": 0.7930917710802574
},
"eval_QTY": {
"f1": 0.9347491965907503,
"number": 7095,
"precision": 0.9267211525141986,
"recall": 0.9429175475687104
},
"eval_RANGE_END": {
"f1": 0.3902439024390244,
"number": 74,
"precision": 0.35555555555555557,
"recall": 0.43243243243243246
},
"eval_UNIT": {
"f1": 0.932516381584546,
"number": 5723,
"precision": 0.9089250165892502,
"recall": 0.9573650183470208
},
"eval_loss": 1.0006664991378784,
"eval_overall_accuracy": 0.7768883964390785,
"eval_overall_f1": 0.7786074142146806,
"eval_overall_precision": 0.733323029366306,
"eval_overall_recall": 0.8298527302620072,
"eval_runtime": 10.2957,
"eval_samples_per_second": 826.172,
"eval_steps_per_second": 25.836,
"step": 12000
},
{
"epoch": 2.47,
"learning_rate": 2.5252425262324292e-05,
"loss": 1.0,
"step": 12500
},
{
"epoch": 2.57,
"learning_rate": 2.4262522272817265e-05,
"loss": 1.0048,
"step": 13000
},
{
"epoch": 2.57,
"eval_COMMENT": {
"f1": 0.542151434082001,
"number": 6836,
"precision": 0.47168994262206343,
"recall": 0.6373610298420129
},
"eval_NAME": {
"f1": 0.7605572485851111,
"number": 8859,
"precision": 0.7342649994746243,
"recall": 0.7888023478947963
},
"eval_QTY": {
"f1": 0.9334144527817211,
"number": 7095,
"precision": 0.8968563263185243,
"recall": 0.9730796335447498
},
"eval_RANGE_END": {
"f1": 0.4028776978417266,
"number": 74,
"precision": 0.4307692307692308,
"recall": 0.3783783783783784
},
"eval_UNIT": {
"f1": 0.9308501753799899,
"number": 5723,
"precision": 0.8915373540233562,
"recall": 0.9737899702952997
},
"eval_loss": 0.9942804574966431,
"eval_overall_accuracy": 0.7787414511137432,
"eval_overall_f1": 0.777442751202021,
"eval_overall_precision": 0.72784423828125,
"eval_overall_recall": 0.8342953090565641,
"eval_runtime": 9.9772,
"eval_samples_per_second": 852.54,
"eval_steps_per_second": 26.661,
"step": 13000
},
{
"epoch": 2.67,
"learning_rate": 2.3272619283310237e-05,
"loss": 0.9991,
"step": 13500
},
{
"epoch": 2.77,
"learning_rate": 2.2282716293803206e-05,
"loss": 0.9911,
"step": 14000
},
{
"epoch": 2.77,
"eval_COMMENT": {
"f1": 0.5497025285076846,
"number": 6836,
"precision": 0.4768817204301075,
"recall": 0.6487712112346401
},
"eval_NAME": {
"f1": 0.7610256410256411,
"number": 8859,
"precision": 0.729257190151045,
"recall": 0.7956880009030365
},
"eval_QTY": {
"f1": 0.9355014287658184,
"number": 7095,
"precision": 0.9042483230303827,
"recall": 0.9689922480620154
},
"eval_RANGE_END": {
"f1": 0.4132231404958678,
"number": 74,
"precision": 0.5319148936170213,
"recall": 0.33783783783783783
},
"eval_UNIT": {
"f1": 0.9322147651006711,
"number": 5723,
"precision": 0.8965628529933839,
"recall": 0.9708195002621003
},
"eval_loss": 0.9951051473617554,
"eval_overall_accuracy": 0.7786077255186643,
"eval_overall_f1": 0.7798045602605863,
"eval_overall_precision": 0.7295888824551245,
"eval_overall_recall": 0.8374435932416833,
"eval_runtime": 11.0903,
"eval_samples_per_second": 766.978,
"eval_steps_per_second": 23.985,
"step": 14000
},
{
"epoch": 2.87,
"learning_rate": 2.129281330429618e-05,
"loss": 0.9912,
"step": 14500
},
{
"epoch": 2.97,
"learning_rate": 2.0302910314789152e-05,
"loss": 0.9991,
"step": 15000
},
{
"epoch": 2.97,
"eval_COMMENT": {
"f1": 0.5583836904107916,
"number": 6836,
"precision": 0.4791033832617576,
"recall": 0.6691047396138092
},
"eval_NAME": {
"f1": 0.7690366469168313,
"number": 8859,
"precision": 0.7481054541573273,
"recall": 0.7911728186025511
},
"eval_QTY": {
"f1": 0.935168616655196,
"number": 7095,
"precision": 0.9137861466039005,
"recall": 0.9575757575757575
},
"eval_RANGE_END": {
"f1": 0.4109589041095891,
"number": 74,
"precision": 0.4166666666666667,
"recall": 0.40540540540540543
},
"eval_UNIT": {
"f1": 0.9313129631171426,
"number": 5723,
"precision": 0.8894720101781171,
"recall": 0.977284640922593
},
"eval_loss": 0.9920729398727417,
"eval_overall_accuracy": 0.7806709204141673,
"eval_overall_f1": 0.783059806192698,
"eval_overall_precision": 0.7336981443551099,
"eval_overall_recall": 0.839542449365096,
"eval_runtime": 9.9635,
"eval_samples_per_second": 853.712,
"eval_steps_per_second": 26.697,
"step": 15000
},
{
"epoch": 3.07,
"learning_rate": 1.9313007325282124e-05,
"loss": 0.9903,
"step": 15500
},
{
"epoch": 3.17,
"learning_rate": 1.8323104335775097e-05,
"loss": 0.9805,
"step": 16000
},
{
"epoch": 3.17,
"eval_COMMENT": {
"f1": 0.5583219220714553,
"number": 6836,
"precision": 0.4859154929577465,
"recall": 0.6560854300760679
},
"eval_NAME": {
"f1": 0.7669049828084921,
"number": 8859,
"precision": 0.7423922231614539,
"recall": 0.7930917710802574
},
"eval_QTY": {
"f1": 0.9368610113290964,
"number": 7095,
"precision": 0.9187102018696653,
"recall": 0.9557434813248766
},
"eval_RANGE_END": {
"f1": 0.4087591240875913,
"number": 74,
"precision": 0.4444444444444444,
"recall": 0.3783783783783784
},
"eval_UNIT": {
"f1": 0.9333669863705198,
"number": 5723,
"precision": 0.9000486775920817,
"recall": 0.9692468984798183
},
"eval_loss": 0.987960696220398,
"eval_overall_accuracy": 0.7821992129293547,
"eval_overall_f1": 0.783963999474445,
"eval_overall_precision": 0.7388935327079657,
"eval_overall_recall": 0.8348899849581978,
"eval_runtime": 9.4297,
"eval_samples_per_second": 902.044,
"eval_steps_per_second": 28.209,
"step": 16000
},
{
"epoch": 3.27,
"learning_rate": 1.7333201346268066e-05,
"loss": 0.9868,
"step": 16500
},
{
"epoch": 3.37,
"learning_rate": 1.634329835676104e-05,
"loss": 0.9848,
"step": 17000
},
{
"epoch": 3.37,
"eval_COMMENT": {
"f1": 0.5623249735515589,
"number": 6836,
"precision": 0.48933174482833314,
"recall": 0.6609128145114102
},
"eval_NAME": {
"f1": 0.7673142355394577,
"number": 8859,
"precision": 0.7466623945316672,
"recall": 0.7891409865673327
},
"eval_QTY": {
"f1": 0.936261013215859,
"number": 7095,
"precision": 0.9149737656397148,
"recall": 0.9585623678646934
},
"eval_RANGE_END": {
"f1": 0.3795620437956204,
"number": 74,
"precision": 0.4126984126984127,
"recall": 0.35135135135135137
},
"eval_UNIT": {
"f1": 0.9321663019693655,
"number": 5723,
"precision": 0.899171943497321,
"recall": 0.9676742966975362
},
"eval_loss": 0.9842170476913452,
"eval_overall_accuracy": 0.7823902494937531,
"eval_overall_f1": 0.7848384233538249,
"eval_overall_precision": 0.7402561160894235,
"eval_overall_recall": 0.8351348515059293,
"eval_runtime": 7.8716,
"eval_samples_per_second": 1080.599,
"eval_steps_per_second": 33.793,
"step": 17000
},
{
"epoch": 3.46,
"learning_rate": 1.535339536725401e-05,
"loss": 0.9841,
"step": 17500
},
{
"epoch": 3.56,
"learning_rate": 1.4363492377746981e-05,
"loss": 0.9771,
"step": 18000
},
{
"epoch": 3.56,
"eval_COMMENT": {
"f1": 0.5630382256365777,
"number": 6836,
"precision": 0.4883396023643203,
"recall": 0.6647162083089526
},
"eval_NAME": {
"f1": 0.7652202248411449,
"number": 8859,
"precision": 0.7373874816830647,
"recall": 0.795236482672988
},
"eval_QTY": {
"f1": 0.9356418569359177,
"number": 7095,
"precision": 0.9162388543636855,
"recall": 0.9558844256518675
},
"eval_RANGE_END": {
"f1": 0.4520547945205479,
"number": 74,
"precision": 0.4583333333333333,
"recall": 0.44594594594594594
},
"eval_UNIT": {
"f1": 0.9327950206072841,
"number": 5723,
"precision": 0.8992864093415505,
"recall": 0.968897431417089
},
"eval_loss": 0.983447253704071,
"eval_overall_accuracy": 0.7836510908187827,
"eval_overall_f1": 0.7841076515077104,
"eval_overall_precision": 0.7369149819994462,
"eval_overall_recall": 0.8377584216601952,
"eval_runtime": 9.9588,
"eval_samples_per_second": 854.123,
"eval_steps_per_second": 26.71,
"step": 18000
},
{
"epoch": 3.66,
"learning_rate": 1.3373589388239954e-05,
"loss": 0.9815,
"step": 18500
},
{
"epoch": 3.76,
"learning_rate": 1.2383686398732925e-05,
"loss": 0.9787,
"step": 19000
},
{
"epoch": 3.76,
"eval_COMMENT": {
"f1": 0.5682919349892671,
"number": 6836,
"precision": 0.4892808110676946,
"recall": 0.677735517846694
},
"eval_NAME": {
"f1": 0.7695169319984682,
"number": 8859,
"precision": 0.7466029723991507,
"recall": 0.7938819279828423
},
"eval_QTY": {
"f1": 0.9372361774271676,
"number": 7095,
"precision": 0.9206090266449157,
"recall": 0.9544749823819592
},
"eval_RANGE_END": {
"f1": 0.4189189189189189,
"number": 74,
"precision": 0.4189189189189189,
"recall": 0.4189189189189189
},
"eval_UNIT": {
"f1": 0.9332317847169331,
"number": 5723,
"precision": 0.9048244174597965,
"recall": 0.9634806919447843
},
"eval_loss": 0.9832035899162292,
"eval_overall_accuracy": 0.7843961334199365,
"eval_overall_f1": 0.786327868852459,
"eval_overall_precision": 0.7399191682349675,
"eval_overall_recall": 0.8389477734634624,
"eval_runtime": 10.5172,
"eval_samples_per_second": 808.772,
"eval_steps_per_second": 25.292,
"step": 19000
},
{
"epoch": 3.86,
"learning_rate": 1.1393783409225896e-05,
"loss": 0.9744,
"step": 19500
},
{
"epoch": 3.96,
"learning_rate": 1.0403880419718868e-05,
"loss": 0.9746,
"step": 20000
},
{
"epoch": 3.96,
"eval_COMMENT": {
"f1": 0.5697782746413266,
"number": 6836,
"precision": 0.4950890447922288,
"recall": 0.6710064365125804
},
"eval_NAME": {
"f1": 0.768101654069449,
"number": 8859,
"precision": 0.7460368124268539,
"recall": 0.7915114572750874
},
"eval_QTY": {
"f1": 0.9370073342929603,
"number": 7095,
"precision": 0.9120629837203096,
"recall": 0.9633544749823819
},
"eval_RANGE_END": {
"f1": 0.43312101910828027,
"number": 74,
"precision": 0.40963855421686746,
"recall": 0.4594594594594595
},
"eval_UNIT": {
"f1": 0.9337932194834694,
"number": 5723,
"precision": 0.9003893575600259,
"recall": 0.9697710990739122
},
"eval_loss": 0.9827048778533936,
"eval_overall_accuracy": 0.7846444809536546,
"eval_overall_f1": 0.7875786988457503,
"eval_overall_precision": 0.7411819163709304,
"eval_overall_recall": 0.8401721062021198,
"eval_runtime": 8.1048,
"eval_samples_per_second": 1049.503,
"eval_steps_per_second": 32.82,
"step": 20000
},
{
"epoch": 4.06,
"learning_rate": 9.41397743021184e-06,
"loss": 0.9818,
"step": 20500
},
{
"epoch": 4.16,
"learning_rate": 8.424074440704812e-06,
"loss": 0.976,
"step": 21000
},
{
"epoch": 4.16,
"eval_COMMENT": {
"f1": 0.5667608401916225,
"number": 6836,
"precision": 0.4884607241160279,
"recall": 0.6749561146869514
},
"eval_NAME": {
"f1": 0.7705115565779385,
"number": 8859,
"precision": 0.7483774869666986,
"recall": 0.7939948075403545
},
"eval_QTY": {
"f1": 0.936799394814662,
"number": 7095,
"precision": 0.9147193123824873,
"recall": 0.9599718111346018
},
"eval_RANGE_END": {
"f1": 0.42857142857142855,
"number": 74,
"precision": 0.4125,
"recall": 0.44594594594594594
},
"eval_UNIT": {
"f1": 0.9335247909451813,
"number": 5723,
"precision": 0.9035317200784827,
"recall": 0.9655774943211602
},
"eval_loss": 0.9835863709449768,
"eval_overall_accuracy": 0.7854086272112483,
"eval_overall_f1": 0.7865212692798899,
"eval_overall_precision": 0.7393111090590082,
"eval_overall_recall": 0.8401721062021198,
"eval_runtime": 10.4319,
"eval_samples_per_second": 815.38,
"eval_steps_per_second": 25.499,
"step": 21000
},
{
"epoch": 4.26,
"learning_rate": 7.434171451197783e-06,
"loss": 0.9902,
"step": 21500
},
{
"epoch": 4.36,
"learning_rate": 6.444268461690754e-06,
"loss": 0.9635,
"step": 22000
},
{
"epoch": 4.36,
"eval_COMMENT": {
"f1": 0.5716760502381983,
"number": 6836,
"precision": 0.49533612093920876,
"recall": 0.6758338209479228
},
"eval_NAME": {
"f1": 0.7705016685814322,
"number": 8859,
"precision": 0.7475583864118895,
"recall": 0.7948978440004515
},
"eval_QTY": {
"f1": 0.9371633752244165,
"number": 7095,
"precision": 0.9186408555570597,
"recall": 0.9564482029598309
},
"eval_RANGE_END": {
"f1": 0.43373493975903615,
"number": 74,
"precision": 0.391304347826087,
"recall": 0.4864864864864865
},
"eval_UNIT": {
"f1": 0.9337832138338253,
"number": 5723,
"precision": 0.9026418786692759,
"recall": 0.9671500961034423
},
"eval_loss": 0.9831692576408386,
"eval_overall_accuracy": 0.7850838650517709,
"eval_overall_f1": 0.7882188858807121,
"eval_overall_precision": 0.7422893874775944,
"eval_overall_recall": 0.84020708713751,
"eval_runtime": 9.9258,
"eval_samples_per_second": 856.957,
"eval_steps_per_second": 26.799,
"step": 22000
},
{
"epoch": 4.45,
"learning_rate": 5.4543654721837265e-06,
"loss": 0.967,
"step": 22500
},
{
"epoch": 4.55,
"learning_rate": 4.4644624826766974e-06,
"loss": 0.9688,
"step": 23000
},
{
"epoch": 4.55,
"eval_COMMENT": {
"f1": 0.5723930522310194,
"number": 6836,
"precision": 0.4930739135032251,
"recall": 0.6821240491515506
},
"eval_NAME": {
"f1": 0.7698282463625424,
"number": 8859,
"precision": 0.7467897697124058,
"recall": 0.7943334462128908
},
"eval_QTY": {
"f1": 0.9373659076752716,
"number": 7095,
"precision": 0.9208593962469405,
"recall": 0.9544749823819592
},
"eval_RANGE_END": {
"f1": 0.4301075268817204,
"number": 74,
"precision": 0.35714285714285715,
"recall": 0.5405405405405406
},
"eval_UNIT": {
"f1": 0.9337379868487607,
"number": 5723,
"precision": 0.9021013194331324,
"recall": 0.9676742966975362
},
"eval_loss": 0.9836147427558899,
"eval_overall_accuracy": 0.7860008405608834,
"eval_overall_f1": 0.7875949698716269,
"eval_overall_precision": 0.7403416961674619,
"eval_overall_recall": 0.8412914961346066,
"eval_runtime": 8.0875,
"eval_samples_per_second": 1051.747,
"eval_steps_per_second": 32.89,
"step": 23000
},
{
"epoch": 4.65,
"learning_rate": 3.4745594931696697e-06,
"loss": 0.9686,
"step": 23500
},
{
"epoch": 4.75,
"learning_rate": 2.484656503662641e-06,
"loss": 0.9669,
"step": 24000
},
{
"epoch": 4.75,
"eval_COMMENT": {
"f1": 0.5733910891089109,
"number": 6836,
"precision": 0.4968897468897469,
"recall": 0.677735517846694
},
"eval_NAME": {
"f1": 0.7695764151460354,
"number": 8859,
"precision": 0.7478168264110756,
"recall": 0.7926402528502088
},
"eval_QTY": {
"f1": 0.9377799214497348,
"number": 7095,
"precision": 0.9173631706659477,
"recall": 0.9591261451726568
},
"eval_RANGE_END": {
"f1": 0.43902439024390244,
"number": 74,
"precision": 0.4,
"recall": 0.4864864864864865
},
"eval_UNIT": {
"f1": 0.9331987211845869,
"number": 5723,
"precision": 0.8998864189518092,
"recall": 0.9690721649484536
},
"eval_loss": 0.9803335070610046,
"eval_overall_accuracy": 0.7873189928552325,
"eval_overall_f1": 0.7886242865577643,
"eval_overall_precision": 0.7423807318202872,
"eval_overall_recall": 0.8410116486514849,
"eval_runtime": 11.2838,
"eval_samples_per_second": 753.825,
"eval_steps_per_second": 23.574,
"step": 24000
},
{
"epoch": 4.85,
"learning_rate": 1.4947535141556129e-06,
"loss": 0.965,
"step": 24500
},
{
"epoch": 4.95,
"learning_rate": 5.048505246485845e-07,
"loss": 0.9691,
"step": 25000
},
{
"epoch": 4.95,
"eval_COMMENT": {
"f1": 0.5725781491798204,
"number": 6836,
"precision": 0.4962978860392746,
"recall": 0.6765652428320655
},
"eval_NAME": {
"f1": 0.7694330320460149,
"number": 8859,
"precision": 0.7474457215836526,
"recall": 0.7927531324077209
},
"eval_QTY": {
"f1": 0.9377026560883062,
"number": 7095,
"precision": 0.9183783783783783,
"recall": 0.9578576462297392
},
"eval_RANGE_END": {
"f1": 0.4484848484848485,
"number": 74,
"precision": 0.4065934065934066,
"recall": 0.5
},
"eval_UNIT": {
"f1": 0.933075174678003,
"number": 5723,
"precision": 0.9002599090318388,
"recall": 0.968373230822995
},
"eval_loss": 0.9796159267425537,
"eval_overall_accuracy": 0.7870897489779544,
"eval_overall_f1": 0.7882984134276199,
"eval_overall_precision": 0.7423212409616217,
"eval_overall_recall": 0.8403470108790709,
"eval_runtime": 8.9053,
"eval_samples_per_second": 955.162,
"eval_steps_per_second": 29.87,
"step": 25000
},
{
"epoch": 5.0,
"step": 25255,
"total_flos": 1840716078790068.0,
"train_loss": 1.016139645029875,
"train_runtime": 1364.1405,
"train_samples_per_second": 592.325,
"train_steps_per_second": 18.513
}
],
"logging_steps": 500,
"max_steps": 25255,
"num_train_epochs": 5,
"save_steps": 1000,
"total_flos": 1840716078790068.0,
"trial_name": null,
"trial_params": null
}