segformer-finetuned-tt-225-2k / trainer_state.json
Saumya-Mundra's picture
End of training
0ff762b verified
raw
history blame
39 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 68.96551724137932,
"eval_steps": 500,
"global_step": 2000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"eval_accuracy_no_text": 0.9411412367275587,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9411412367275587,
"eval_iou_text": 0.0,
"eval_loss": 0.26036718487739563,
"eval_mean_accuracy": 0.9411412367275587,
"eval_mean_iou": 0.47057061836377934,
"eval_overall_accuracy": 0.9411412367275587,
"eval_runtime": 2.53,
"eval_samples_per_second": 29.644,
"eval_steps_per_second": 3.953,
"step": 29
},
{
"epoch": 2.0,
"eval_accuracy_no_text": 0.9556761981853922,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9556761981853922,
"eval_iou_text": 0.0,
"eval_loss": 0.20939157903194427,
"eval_mean_accuracy": 0.9556761981853922,
"eval_mean_iou": 0.4778380990926961,
"eval_overall_accuracy": 0.9556761981853922,
"eval_runtime": 2.5445,
"eval_samples_per_second": 29.475,
"eval_steps_per_second": 3.93,
"step": 58
},
{
"epoch": 3.0,
"eval_accuracy_no_text": 0.9646688438230138,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9646688438230138,
"eval_iou_text": 0.0,
"eval_loss": 0.18179728090763092,
"eval_mean_accuracy": 0.9646688438230138,
"eval_mean_iou": 0.4823344219115069,
"eval_overall_accuracy": 0.9646688438230138,
"eval_runtime": 2.5227,
"eval_samples_per_second": 29.73,
"eval_steps_per_second": 3.964,
"step": 87
},
{
"epoch": 3.4482758620689653,
"grad_norm": 69.57986450195312,
"learning_rate": 5.7005e-05,
"loss": 0.323,
"step": 100
},
{
"epoch": 4.0,
"eval_accuracy_no_text": 0.9392420264366191,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9392420264366191,
"eval_iou_text": 0.0,
"eval_loss": 0.20981624722480774,
"eval_mean_accuracy": 0.9392420264366191,
"eval_mean_iou": 0.46962101321830957,
"eval_overall_accuracy": 0.9392420264366191,
"eval_runtime": 2.5287,
"eval_samples_per_second": 29.66,
"eval_steps_per_second": 3.955,
"step": 116
},
{
"epoch": 5.0,
"eval_accuracy_no_text": 0.9698847328744703,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9698847328744703,
"eval_iou_text": 0.0,
"eval_loss": 0.17170381546020508,
"eval_mean_accuracy": 0.9698847328744703,
"eval_mean_iou": 0.48494236643723515,
"eval_overall_accuracy": 0.9698847328744703,
"eval_runtime": 2.4809,
"eval_samples_per_second": 30.231,
"eval_steps_per_second": 4.031,
"step": 145
},
{
"epoch": 6.0,
"eval_accuracy_no_text": 0.9680704296500666,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9680704296500666,
"eval_iou_text": 0.0,
"eval_loss": 0.14835301041603088,
"eval_mean_accuracy": 0.9680704296500666,
"eval_mean_iou": 0.4840352148250333,
"eval_overall_accuracy": 0.9680704296500666,
"eval_runtime": 2.518,
"eval_samples_per_second": 29.786,
"eval_steps_per_second": 3.971,
"step": 174
},
{
"epoch": 6.896551724137931,
"grad_norm": 2.1472854614257812,
"learning_rate": 5.401e-05,
"loss": 0.2027,
"step": 200
},
{
"epoch": 7.0,
"eval_accuracy_no_text": 0.9630674570564881,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9630674570564881,
"eval_iou_text": 0.0,
"eval_loss": 0.16156023740768433,
"eval_mean_accuracy": 0.9630674570564881,
"eval_mean_iou": 0.48153372852824405,
"eval_overall_accuracy": 0.9630674570564881,
"eval_runtime": 2.5744,
"eval_samples_per_second": 29.133,
"eval_steps_per_second": 3.884,
"step": 203
},
{
"epoch": 8.0,
"eval_accuracy_no_text": 0.9727854516339448,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9727854516339448,
"eval_iou_text": 0.0,
"eval_loss": 0.15030205249786377,
"eval_mean_accuracy": 0.9727854516339448,
"eval_mean_iou": 0.4863927258169724,
"eval_overall_accuracy": 0.9727854516339448,
"eval_runtime": 2.6278,
"eval_samples_per_second": 28.541,
"eval_steps_per_second": 3.806,
"step": 232
},
{
"epoch": 9.0,
"eval_accuracy_no_text": 0.9566264239140725,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9566264239140725,
"eval_iou_text": 0.0,
"eval_loss": 0.16279229521751404,
"eval_mean_accuracy": 0.9566264239140725,
"eval_mean_iou": 0.47831321195703624,
"eval_overall_accuracy": 0.9566264239140725,
"eval_runtime": 2.4976,
"eval_samples_per_second": 30.029,
"eval_steps_per_second": 4.004,
"step": 261
},
{
"epoch": 10.0,
"eval_accuracy_no_text": 0.9670189360247539,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9670189360247539,
"eval_iou_text": 0.0,
"eval_loss": 0.142441064119339,
"eval_mean_accuracy": 0.9670189360247539,
"eval_mean_iou": 0.48350946801237693,
"eval_overall_accuracy": 0.9670189360247539,
"eval_runtime": 2.5057,
"eval_samples_per_second": 29.931,
"eval_steps_per_second": 3.991,
"step": 290
},
{
"epoch": 10.344827586206897,
"grad_norm": 0.8168253302574158,
"learning_rate": 5.1015e-05,
"loss": 0.1693,
"step": 300
},
{
"epoch": 11.0,
"eval_accuracy_no_text": 0.9806479024118233,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9806479024118233,
"eval_iou_text": 0.0,
"eval_loss": 0.12359502166509628,
"eval_mean_accuracy": 0.9806479024118233,
"eval_mean_iou": 0.49032395120591166,
"eval_overall_accuracy": 0.9806479024118233,
"eval_runtime": 2.5703,
"eval_samples_per_second": 29.18,
"eval_steps_per_second": 3.891,
"step": 319
},
{
"epoch": 12.0,
"eval_accuracy_no_text": 0.9632355222732392,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9632355222732392,
"eval_iou_text": 0.0,
"eval_loss": 0.13876928389072418,
"eval_mean_accuracy": 0.9632355222732392,
"eval_mean_iou": 0.4816177611366196,
"eval_overall_accuracy": 0.9632355222732392,
"eval_runtime": 2.4816,
"eval_samples_per_second": 30.223,
"eval_steps_per_second": 4.03,
"step": 348
},
{
"epoch": 13.0,
"eval_accuracy_no_text": 0.9576819795385817,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9576819795385817,
"eval_iou_text": 0.0,
"eval_loss": 0.1554112434387207,
"eval_mean_accuracy": 0.9576819795385817,
"eval_mean_iou": 0.47884098976929085,
"eval_overall_accuracy": 0.9576819795385817,
"eval_runtime": 2.4981,
"eval_samples_per_second": 30.023,
"eval_steps_per_second": 4.003,
"step": 377
},
{
"epoch": 13.793103448275861,
"grad_norm": 3.8616011142730713,
"learning_rate": 4.8020000000000004e-05,
"loss": 0.1502,
"step": 400
},
{
"epoch": 14.0,
"eval_accuracy_no_text": 0.9521418075512791,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9521418075512791,
"eval_iou_text": 0.0,
"eval_loss": 0.15367722511291504,
"eval_mean_accuracy": 0.9521418075512791,
"eval_mean_iou": 0.47607090377563954,
"eval_overall_accuracy": 0.9521418075512791,
"eval_runtime": 2.5218,
"eval_samples_per_second": 29.741,
"eval_steps_per_second": 3.965,
"step": 406
},
{
"epoch": 15.0,
"eval_accuracy_no_text": 0.9562213523275425,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9562213523275425,
"eval_iou_text": 0.0,
"eval_loss": 0.14778192341327667,
"eval_mean_accuracy": 0.9562213523275425,
"eval_mean_iou": 0.47811067616377123,
"eval_overall_accuracy": 0.9562213523275425,
"eval_runtime": 2.4931,
"eval_samples_per_second": 30.083,
"eval_steps_per_second": 4.011,
"step": 435
},
{
"epoch": 16.0,
"eval_accuracy_no_text": 0.9650662991610505,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9650662991610505,
"eval_iou_text": 0.0,
"eval_loss": 0.1367165446281433,
"eval_mean_accuracy": 0.9650662991610505,
"eval_mean_iou": 0.48253314958052523,
"eval_overall_accuracy": 0.9650662991610505,
"eval_runtime": 2.5189,
"eval_samples_per_second": 29.774,
"eval_steps_per_second": 3.97,
"step": 464
},
{
"epoch": 17.0,
"eval_accuracy_no_text": 0.9702622167164582,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9702622167164582,
"eval_iou_text": 0.0,
"eval_loss": 0.13298067450523376,
"eval_mean_accuracy": 0.9702622167164582,
"eval_mean_iou": 0.4851311083582291,
"eval_overall_accuracy": 0.9702622167164582,
"eval_runtime": 2.4829,
"eval_samples_per_second": 30.207,
"eval_steps_per_second": 4.028,
"step": 493
},
{
"epoch": 17.24137931034483,
"grad_norm": 0.5257724523544312,
"learning_rate": 4.5025000000000003e-05,
"loss": 0.127,
"step": 500
},
{
"epoch": 18.0,
"eval_accuracy_no_text": 0.9695612961884569,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9695612961884569,
"eval_iou_text": 0.0,
"eval_loss": 0.13220080733299255,
"eval_mean_accuracy": 0.9695612961884569,
"eval_mean_iou": 0.48478064809422844,
"eval_overall_accuracy": 0.9695612961884569,
"eval_runtime": 2.4852,
"eval_samples_per_second": 30.179,
"eval_steps_per_second": 4.024,
"step": 522
},
{
"epoch": 19.0,
"eval_accuracy_no_text": 0.9684219618305219,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9684219618305219,
"eval_iou_text": 0.0,
"eval_loss": 0.13408276438713074,
"eval_mean_accuracy": 0.9684219618305219,
"eval_mean_iou": 0.48421098091526094,
"eval_overall_accuracy": 0.9684219618305219,
"eval_runtime": 2.5013,
"eval_samples_per_second": 29.984,
"eval_steps_per_second": 3.998,
"step": 551
},
{
"epoch": 20.0,
"eval_accuracy_no_text": 0.9764442974101032,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9764442974101032,
"eval_iou_text": 0.0,
"eval_loss": 0.12438876926898956,
"eval_mean_accuracy": 0.9764442974101032,
"eval_mean_iou": 0.4882221487050516,
"eval_overall_accuracy": 0.9764442974101032,
"eval_runtime": 2.4814,
"eval_samples_per_second": 30.225,
"eval_steps_per_second": 4.03,
"step": 580
},
{
"epoch": 20.689655172413794,
"grad_norm": 0.9491066336631775,
"learning_rate": 4.203e-05,
"loss": 0.1173,
"step": 600
},
{
"epoch": 21.0,
"eval_accuracy_no_text": 0.9792908561802923,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9792908561802923,
"eval_iou_text": 0.0,
"eval_loss": 0.1200159564614296,
"eval_mean_accuracy": 0.9792908561802923,
"eval_mean_iou": 0.48964542809014616,
"eval_overall_accuracy": 0.9792908561802923,
"eval_runtime": 2.4555,
"eval_samples_per_second": 30.544,
"eval_steps_per_second": 4.072,
"step": 609
},
{
"epoch": 22.0,
"eval_accuracy_no_text": 0.9652650550383967,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9652650550383967,
"eval_iou_text": 0.0,
"eval_loss": 0.13735905289649963,
"eval_mean_accuracy": 0.9652650550383967,
"eval_mean_iou": 0.48263252751919833,
"eval_overall_accuracy": 0.9652650550383967,
"eval_runtime": 2.4618,
"eval_samples_per_second": 30.465,
"eval_steps_per_second": 4.062,
"step": 638
},
{
"epoch": 23.0,
"eval_accuracy_no_text": 0.9768179449195166,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9768179449195166,
"eval_iou_text": 0.0,
"eval_loss": 0.1247735321521759,
"eval_mean_accuracy": 0.9768179449195166,
"eval_mean_iou": 0.4884089724597583,
"eval_overall_accuracy": 0.9768179449195166,
"eval_runtime": 2.5434,
"eval_samples_per_second": 29.489,
"eval_steps_per_second": 3.932,
"step": 667
},
{
"epoch": 24.0,
"eval_accuracy_no_text": 0.9713726093301188,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9713726093301188,
"eval_iou_text": 0.0,
"eval_loss": 0.12621533870697021,
"eval_mean_accuracy": 0.9713726093301188,
"eval_mean_iou": 0.4856863046650594,
"eval_overall_accuracy": 0.9713726093301188,
"eval_runtime": 2.5006,
"eval_samples_per_second": 29.993,
"eval_steps_per_second": 3.999,
"step": 696
},
{
"epoch": 24.137931034482758,
"grad_norm": 1.288313627243042,
"learning_rate": 3.9035e-05,
"loss": 0.1193,
"step": 700
},
{
"epoch": 25.0,
"eval_accuracy_no_text": 0.9719735595445641,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9719735595445641,
"eval_iou_text": 0.0,
"eval_loss": 0.12352190911769867,
"eval_mean_accuracy": 0.9719735595445641,
"eval_mean_iou": 0.48598677977228205,
"eval_overall_accuracy": 0.9719735595445641,
"eval_runtime": 2.4888,
"eval_samples_per_second": 30.135,
"eval_steps_per_second": 4.018,
"step": 725
},
{
"epoch": 26.0,
"eval_accuracy_no_text": 0.9676620294808619,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9676620294808619,
"eval_iou_text": 0.0,
"eval_loss": 0.1323447823524475,
"eval_mean_accuracy": 0.9676620294808619,
"eval_mean_iou": 0.48383101474043094,
"eval_overall_accuracy": 0.9676620294808619,
"eval_runtime": 2.5267,
"eval_samples_per_second": 29.683,
"eval_steps_per_second": 3.958,
"step": 754
},
{
"epoch": 27.0,
"eval_accuracy_no_text": 0.970691829548133,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.970691829548133,
"eval_iou_text": 0.0,
"eval_loss": 0.12354418635368347,
"eval_mean_accuracy": 0.970691829548133,
"eval_mean_iou": 0.4853459147740665,
"eval_overall_accuracy": 0.970691829548133,
"eval_runtime": 2.5317,
"eval_samples_per_second": 29.625,
"eval_steps_per_second": 3.95,
"step": 783
},
{
"epoch": 27.586206896551722,
"grad_norm": 0.8687916398048401,
"learning_rate": 3.604e-05,
"loss": 0.0912,
"step": 800
},
{
"epoch": 28.0,
"eval_accuracy_no_text": 0.9631846908666286,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9631846908666286,
"eval_iou_text": 0.0,
"eval_loss": 0.13487361371517181,
"eval_mean_accuracy": 0.9631846908666286,
"eval_mean_iou": 0.4815923454333143,
"eval_overall_accuracy": 0.9631846908666286,
"eval_runtime": 2.5216,
"eval_samples_per_second": 29.743,
"eval_steps_per_second": 3.966,
"step": 812
},
{
"epoch": 29.0,
"eval_accuracy_no_text": 0.9631912351986673,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9631912351986673,
"eval_iou_text": 0.0,
"eval_loss": 0.140753835439682,
"eval_mean_accuracy": 0.9631912351986673,
"eval_mean_iou": 0.48159561759933367,
"eval_overall_accuracy": 0.9631912351986673,
"eval_runtime": 2.4836,
"eval_samples_per_second": 30.198,
"eval_steps_per_second": 4.026,
"step": 841
},
{
"epoch": 30.0,
"eval_accuracy_no_text": 0.9753996302001064,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9753996302001064,
"eval_iou_text": 0.0,
"eval_loss": 0.1205972284078598,
"eval_mean_accuracy": 0.9753996302001064,
"eval_mean_iou": 0.4876998151000532,
"eval_overall_accuracy": 0.9753996302001064,
"eval_runtime": 2.4858,
"eval_samples_per_second": 30.171,
"eval_steps_per_second": 4.023,
"step": 870
},
{
"epoch": 31.0,
"eval_accuracy_no_text": 0.9646436819946582,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9646436819946582,
"eval_iou_text": 0.0,
"eval_loss": 0.13470463454723358,
"eval_mean_accuracy": 0.9646436819946582,
"eval_mean_iou": 0.4823218409973291,
"eval_overall_accuracy": 0.9646436819946582,
"eval_runtime": 2.4993,
"eval_samples_per_second": 30.009,
"eval_steps_per_second": 4.001,
"step": 899
},
{
"epoch": 31.03448275862069,
"grad_norm": 1.7799224853515625,
"learning_rate": 3.3045000000000006e-05,
"loss": 0.1005,
"step": 900
},
{
"epoch": 32.0,
"eval_accuracy_no_text": 0.9757926850390138,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9757926850390138,
"eval_iou_text": 0.0,
"eval_loss": 0.11876140534877777,
"eval_mean_accuracy": 0.9757926850390138,
"eval_mean_iou": 0.4878963425195069,
"eval_overall_accuracy": 0.9757926850390138,
"eval_runtime": 2.4793,
"eval_samples_per_second": 30.25,
"eval_steps_per_second": 4.033,
"step": 928
},
{
"epoch": 33.0,
"eval_accuracy_no_text": 0.9781089272474506,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9781089272474506,
"eval_iou_text": 0.0,
"eval_loss": 0.12336363643407822,
"eval_mean_accuracy": 0.9781089272474506,
"eval_mean_iou": 0.4890544636237253,
"eval_overall_accuracy": 0.9781089272474506,
"eval_runtime": 2.4798,
"eval_samples_per_second": 30.245,
"eval_steps_per_second": 4.033,
"step": 957
},
{
"epoch": 34.0,
"eval_accuracy_no_text": 0.9848704069931378,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9848704069931378,
"eval_iou_text": 0.0,
"eval_loss": 0.12059501558542252,
"eval_mean_accuracy": 0.9848704069931378,
"eval_mean_iou": 0.4924352034965689,
"eval_overall_accuracy": 0.9848704069931378,
"eval_runtime": 2.5063,
"eval_samples_per_second": 29.925,
"eval_steps_per_second": 3.99,
"step": 986
},
{
"epoch": 34.48275862068966,
"grad_norm": 1.1413416862487793,
"learning_rate": 3.0050000000000002e-05,
"loss": 0.0933,
"step": 1000
},
{
"epoch": 35.0,
"eval_accuracy_no_text": 0.9789616085787618,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9789616085787618,
"eval_iou_text": 0.0,
"eval_loss": 0.12342917174100876,
"eval_mean_accuracy": 0.9789616085787618,
"eval_mean_iou": 0.4894808042893809,
"eval_overall_accuracy": 0.9789616085787618,
"eval_runtime": 2.4727,
"eval_samples_per_second": 30.332,
"eval_steps_per_second": 4.044,
"step": 1015
},
{
"epoch": 36.0,
"eval_accuracy_no_text": 0.9753304069638011,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9753304069638011,
"eval_iou_text": 0.0,
"eval_loss": 0.1192336305975914,
"eval_mean_accuracy": 0.9753304069638011,
"eval_mean_iou": 0.48766520348190057,
"eval_overall_accuracy": 0.9753304069638011,
"eval_runtime": 2.4878,
"eval_samples_per_second": 30.147,
"eval_steps_per_second": 4.02,
"step": 1044
},
{
"epoch": 37.0,
"eval_accuracy_no_text": 0.9614615005408664,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9614615005408664,
"eval_iou_text": 0.0,
"eval_loss": 0.13877929747104645,
"eval_mean_accuracy": 0.9614615005408664,
"eval_mean_iou": 0.4807307502704332,
"eval_overall_accuracy": 0.9614615005408664,
"eval_runtime": 2.4791,
"eval_samples_per_second": 30.253,
"eval_steps_per_second": 4.034,
"step": 1073
},
{
"epoch": 37.93103448275862,
"grad_norm": 0.8322708606719971,
"learning_rate": 2.7054999999999998e-05,
"loss": 0.0937,
"step": 1100
},
{
"epoch": 38.0,
"eval_accuracy_no_text": 0.9719491311327302,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9719491311327302,
"eval_iou_text": 0.0,
"eval_loss": 0.12549874186515808,
"eval_mean_accuracy": 0.9719491311327302,
"eval_mean_iou": 0.4859745655663651,
"eval_overall_accuracy": 0.9719491311327302,
"eval_runtime": 2.5079,
"eval_samples_per_second": 29.905,
"eval_steps_per_second": 3.987,
"step": 1102
},
{
"epoch": 39.0,
"eval_accuracy_no_text": 0.9754536209394253,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9754536209394253,
"eval_iou_text": 0.0,
"eval_loss": 0.11986999213695526,
"eval_mean_accuracy": 0.9754536209394253,
"eval_mean_iou": 0.48772681046971267,
"eval_overall_accuracy": 0.9754536209394253,
"eval_runtime": 2.5348,
"eval_samples_per_second": 29.588,
"eval_steps_per_second": 3.945,
"step": 1131
},
{
"epoch": 40.0,
"eval_accuracy_no_text": 0.9798514684860512,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9798514684860512,
"eval_iou_text": 0.0,
"eval_loss": 0.11944077908992767,
"eval_mean_accuracy": 0.9798514684860512,
"eval_mean_iou": 0.4899257342430256,
"eval_overall_accuracy": 0.9798514684860512,
"eval_runtime": 2.5215,
"eval_samples_per_second": 29.744,
"eval_steps_per_second": 3.966,
"step": 1160
},
{
"epoch": 41.0,
"eval_accuracy_no_text": 0.9687282478532616,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9687282478532616,
"eval_iou_text": 0.0,
"eval_loss": 0.12739954888820648,
"eval_mean_accuracy": 0.9687282478532616,
"eval_mean_iou": 0.4843641239266308,
"eval_overall_accuracy": 0.9687282478532616,
"eval_runtime": 2.483,
"eval_samples_per_second": 30.206,
"eval_steps_per_second": 4.027,
"step": 1189
},
{
"epoch": 41.37931034482759,
"grad_norm": 0.581297755241394,
"learning_rate": 2.406e-05,
"loss": 0.0923,
"step": 1200
},
{
"epoch": 42.0,
"eval_accuracy_no_text": 0.9703110735401259,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9703110735401259,
"eval_iou_text": 0.0,
"eval_loss": 0.13015305995941162,
"eval_mean_accuracy": 0.9703110735401259,
"eval_mean_iou": 0.48515553677006296,
"eval_overall_accuracy": 0.9703110735401259,
"eval_runtime": 2.4521,
"eval_samples_per_second": 30.586,
"eval_steps_per_second": 4.078,
"step": 1218
},
{
"epoch": 43.0,
"eval_accuracy_no_text": 0.9708830820102969,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9708830820102969,
"eval_iou_text": 0.0,
"eval_loss": 0.1296657919883728,
"eval_mean_accuracy": 0.9708830820102969,
"eval_mean_iou": 0.48544154100514847,
"eval_overall_accuracy": 0.9708830820102969,
"eval_runtime": 2.5553,
"eval_samples_per_second": 29.351,
"eval_steps_per_second": 3.913,
"step": 1247
},
{
"epoch": 44.0,
"eval_accuracy_no_text": 0.9700965774158937,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9700965774158937,
"eval_iou_text": 0.0,
"eval_loss": 0.12993517518043518,
"eval_mean_accuracy": 0.9700965774158937,
"eval_mean_iou": 0.48504828870794686,
"eval_overall_accuracy": 0.9700965774158937,
"eval_runtime": 2.572,
"eval_samples_per_second": 29.16,
"eval_steps_per_second": 3.888,
"step": 1276
},
{
"epoch": 44.827586206896555,
"grad_norm": 0.5657644867897034,
"learning_rate": 2.1065e-05,
"loss": 0.0812,
"step": 1300
},
{
"epoch": 45.0,
"eval_accuracy_no_text": 0.9733241742773675,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9733241742773675,
"eval_iou_text": 0.0,
"eval_loss": 0.12587976455688477,
"eval_mean_accuracy": 0.9733241742773675,
"eval_mean_iou": 0.48666208713868375,
"eval_overall_accuracy": 0.9733241742773675,
"eval_runtime": 2.476,
"eval_samples_per_second": 30.291,
"eval_steps_per_second": 4.039,
"step": 1305
},
{
"epoch": 46.0,
"eval_accuracy_no_text": 0.9766478487031672,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9766478487031672,
"eval_iou_text": 0.0,
"eval_loss": 0.12515687942504883,
"eval_mean_accuracy": 0.9766478487031672,
"eval_mean_iou": 0.4883239243515836,
"eval_overall_accuracy": 0.9766478487031672,
"eval_runtime": 2.468,
"eval_samples_per_second": 30.389,
"eval_steps_per_second": 4.052,
"step": 1334
},
{
"epoch": 47.0,
"eval_accuracy_no_text": 0.9761687020479585,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9761687020479585,
"eval_iou_text": 0.0,
"eval_loss": 0.12227589637041092,
"eval_mean_accuracy": 0.9761687020479585,
"eval_mean_iou": 0.48808435102397923,
"eval_overall_accuracy": 0.9761687020479585,
"eval_runtime": 2.4339,
"eval_samples_per_second": 30.815,
"eval_steps_per_second": 4.109,
"step": 1363
},
{
"epoch": 48.0,
"eval_accuracy_no_text": 0.9757235746360194,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9757235746360194,
"eval_iou_text": 0.0,
"eval_loss": 0.12267401069402695,
"eval_mean_accuracy": 0.9757235746360194,
"eval_mean_iou": 0.4878617873180097,
"eval_overall_accuracy": 0.9757235746360194,
"eval_runtime": 2.4543,
"eval_samples_per_second": 30.559,
"eval_steps_per_second": 4.075,
"step": 1392
},
{
"epoch": 48.275862068965516,
"grad_norm": 1.296958565711975,
"learning_rate": 1.807e-05,
"loss": 0.0901,
"step": 1400
},
{
"epoch": 49.0,
"eval_accuracy_no_text": 0.9759114985155085,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9759114985155085,
"eval_iou_text": 0.0,
"eval_loss": 0.12235772609710693,
"eval_mean_accuracy": 0.9759114985155085,
"eval_mean_iou": 0.48795574925775426,
"eval_overall_accuracy": 0.9759114985155085,
"eval_runtime": 2.4545,
"eval_samples_per_second": 30.556,
"eval_steps_per_second": 4.074,
"step": 1421
},
{
"epoch": 50.0,
"eval_accuracy_no_text": 0.9636262076126153,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9636262076126153,
"eval_iou_text": 0.0,
"eval_loss": 0.13496458530426025,
"eval_mean_accuracy": 0.9636262076126153,
"eval_mean_iou": 0.48181310380630765,
"eval_overall_accuracy": 0.9636262076126153,
"eval_runtime": 2.4536,
"eval_samples_per_second": 30.567,
"eval_steps_per_second": 4.076,
"step": 1450
},
{
"epoch": 51.0,
"eval_accuracy_no_text": 0.9718740969809078,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9718740969809078,
"eval_iou_text": 0.0,
"eval_loss": 0.12845410406589508,
"eval_mean_accuracy": 0.9718740969809078,
"eval_mean_iou": 0.4859370484904539,
"eval_overall_accuracy": 0.9718740969809078,
"eval_runtime": 2.4972,
"eval_samples_per_second": 30.034,
"eval_steps_per_second": 4.005,
"step": 1479
},
{
"epoch": 51.724137931034484,
"grad_norm": 1.6991603374481201,
"learning_rate": 1.5075000000000002e-05,
"loss": 0.083,
"step": 1500
},
{
"epoch": 52.0,
"eval_accuracy_no_text": 0.9695387295262546,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9695387295262546,
"eval_iou_text": 0.0,
"eval_loss": 0.12856803834438324,
"eval_mean_accuracy": 0.9695387295262546,
"eval_mean_iou": 0.4847693647631273,
"eval_overall_accuracy": 0.9695387295262546,
"eval_runtime": 2.5249,
"eval_samples_per_second": 29.704,
"eval_steps_per_second": 3.961,
"step": 1508
},
{
"epoch": 53.0,
"eval_accuracy_no_text": 0.9700392580939001,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9700392580939001,
"eval_iou_text": 0.0,
"eval_loss": 0.12850378453731537,
"eval_mean_accuracy": 0.9700392580939001,
"eval_mean_iou": 0.48501962904695006,
"eval_overall_accuracy": 0.9700392580939001,
"eval_runtime": 2.4685,
"eval_samples_per_second": 30.383,
"eval_steps_per_second": 4.051,
"step": 1537
},
{
"epoch": 54.0,
"eval_accuracy_no_text": 0.9775246199463794,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9775246199463794,
"eval_iou_text": 0.0,
"eval_loss": 0.12516801059246063,
"eval_mean_accuracy": 0.9775246199463794,
"eval_mean_iou": 0.4887623099731897,
"eval_overall_accuracy": 0.9775246199463794,
"eval_runtime": 2.4581,
"eval_samples_per_second": 30.511,
"eval_steps_per_second": 4.068,
"step": 1566
},
{
"epoch": 55.0,
"eval_accuracy_no_text": 0.9734067118443722,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9734067118443722,
"eval_iou_text": 0.0,
"eval_loss": 0.1259094923734665,
"eval_mean_accuracy": 0.9734067118443722,
"eval_mean_iou": 0.4867033559221861,
"eval_overall_accuracy": 0.9734067118443722,
"eval_runtime": 2.4476,
"eval_samples_per_second": 30.642,
"eval_steps_per_second": 4.086,
"step": 1595
},
{
"epoch": 55.172413793103445,
"grad_norm": 0.6328563690185547,
"learning_rate": 1.2079999999999998e-05,
"loss": 0.0822,
"step": 1600
},
{
"epoch": 56.0,
"eval_accuracy_no_text": 0.9642389489080613,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9642389489080613,
"eval_iou_text": 0.0,
"eval_loss": 0.13731960952281952,
"eval_mean_accuracy": 0.9642389489080613,
"eval_mean_iou": 0.48211947445403064,
"eval_overall_accuracy": 0.9642389489080613,
"eval_runtime": 2.4564,
"eval_samples_per_second": 30.532,
"eval_steps_per_second": 4.071,
"step": 1624
},
{
"epoch": 57.0,
"eval_accuracy_no_text": 0.9719260003039729,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9719260003039729,
"eval_iou_text": 0.0,
"eval_loss": 0.12484534084796906,
"eval_mean_accuracy": 0.9719260003039729,
"eval_mean_iou": 0.48596300015198646,
"eval_overall_accuracy": 0.9719260003039729,
"eval_runtime": 2.4497,
"eval_samples_per_second": 30.616,
"eval_steps_per_second": 4.082,
"step": 1653
},
{
"epoch": 58.0,
"eval_accuracy_no_text": 0.9761464174690337,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9761464174690337,
"eval_iou_text": 0.0,
"eval_loss": 0.12278713285923004,
"eval_mean_accuracy": 0.9761464174690337,
"eval_mean_iou": 0.4880732087345169,
"eval_overall_accuracy": 0.9761464174690337,
"eval_runtime": 2.4362,
"eval_samples_per_second": 30.786,
"eval_steps_per_second": 4.105,
"step": 1682
},
{
"epoch": 58.62068965517241,
"grad_norm": 1.0782545804977417,
"learning_rate": 9.085000000000002e-06,
"loss": 0.0777,
"step": 1700
},
{
"epoch": 59.0,
"eval_accuracy_no_text": 0.967698982390218,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.967698982390218,
"eval_iou_text": 0.0,
"eval_loss": 0.13151277601718903,
"eval_mean_accuracy": 0.967698982390218,
"eval_mean_iou": 0.483849491195109,
"eval_overall_accuracy": 0.967698982390218,
"eval_runtime": 2.4544,
"eval_samples_per_second": 30.557,
"eval_steps_per_second": 4.074,
"step": 1711
},
{
"epoch": 60.0,
"eval_accuracy_no_text": 0.9766246614577545,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9766246614577545,
"eval_iou_text": 0.0,
"eval_loss": 0.12540759146213531,
"eval_mean_accuracy": 0.9766246614577545,
"eval_mean_iou": 0.48831233072887725,
"eval_overall_accuracy": 0.9766246614577545,
"eval_runtime": 2.449,
"eval_samples_per_second": 30.624,
"eval_steps_per_second": 4.083,
"step": 1740
},
{
"epoch": 61.0,
"eval_accuracy_no_text": 0.9682487062814643,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9682487062814643,
"eval_iou_text": 0.0,
"eval_loss": 0.13411304354667664,
"eval_mean_accuracy": 0.9682487062814643,
"eval_mean_iou": 0.48412435314073216,
"eval_overall_accuracy": 0.9682487062814643,
"eval_runtime": 2.4399,
"eval_samples_per_second": 30.739,
"eval_steps_per_second": 4.099,
"step": 1769
},
{
"epoch": 62.0,
"eval_accuracy_no_text": 0.9726401787460179,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9726401787460179,
"eval_iou_text": 0.0,
"eval_loss": 0.12687736749649048,
"eval_mean_accuracy": 0.9726401787460179,
"eval_mean_iou": 0.48632008937300897,
"eval_overall_accuracy": 0.9726401787460179,
"eval_runtime": 2.4427,
"eval_samples_per_second": 30.703,
"eval_steps_per_second": 4.094,
"step": 1798
},
{
"epoch": 62.06896551724138,
"grad_norm": 0.44511547684669495,
"learning_rate": 6.0899999999999984e-06,
"loss": 0.079,
"step": 1800
},
{
"epoch": 63.0,
"eval_accuracy_no_text": 0.9755465955876985,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9755465955876985,
"eval_iou_text": 0.0,
"eval_loss": 0.12655200064182281,
"eval_mean_accuracy": 0.9755465955876985,
"eval_mean_iou": 0.48777329779384926,
"eval_overall_accuracy": 0.9755465955876985,
"eval_runtime": 2.433,
"eval_samples_per_second": 30.826,
"eval_steps_per_second": 4.11,
"step": 1827
},
{
"epoch": 64.0,
"eval_accuracy_no_text": 0.9705849764026055,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9705849764026055,
"eval_iou_text": 0.0,
"eval_loss": 0.13036541640758514,
"eval_mean_accuracy": 0.9705849764026055,
"eval_mean_iou": 0.4852924882013028,
"eval_overall_accuracy": 0.9705849764026055,
"eval_runtime": 2.4456,
"eval_samples_per_second": 30.668,
"eval_steps_per_second": 4.089,
"step": 1856
},
{
"epoch": 65.0,
"eval_accuracy_no_text": 0.974686523674516,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.974686523674516,
"eval_iou_text": 0.0,
"eval_loss": 0.12534736096858978,
"eval_mean_accuracy": 0.974686523674516,
"eval_mean_iou": 0.487343261837258,
"eval_overall_accuracy": 0.974686523674516,
"eval_runtime": 2.4366,
"eval_samples_per_second": 30.78,
"eval_steps_per_second": 4.104,
"step": 1885
},
{
"epoch": 65.51724137931035,
"grad_norm": 0.8290705680847168,
"learning_rate": 3.0950000000000026e-06,
"loss": 0.0781,
"step": 1900
},
{
"epoch": 66.0,
"eval_accuracy_no_text": 0.9731443179796158,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9731443179796158,
"eval_iou_text": 0.0,
"eval_loss": 0.12833286821842194,
"eval_mean_accuracy": 0.9731443179796158,
"eval_mean_iou": 0.4865721589898079,
"eval_overall_accuracy": 0.9731443179796158,
"eval_runtime": 2.4332,
"eval_samples_per_second": 30.823,
"eval_steps_per_second": 4.11,
"step": 1914
},
{
"epoch": 67.0,
"eval_accuracy_no_text": 0.9784451705142637,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9784451705142637,
"eval_iou_text": 0.0,
"eval_loss": 0.12897755205631256,
"eval_mean_accuracy": 0.9784451705142637,
"eval_mean_iou": 0.48922258525713186,
"eval_overall_accuracy": 0.9784451705142637,
"eval_runtime": 2.443,
"eval_samples_per_second": 30.7,
"eval_steps_per_second": 4.093,
"step": 1943
},
{
"epoch": 68.0,
"eval_accuracy_no_text": 0.9669106724628387,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9669106724628387,
"eval_iou_text": 0.0,
"eval_loss": 0.13632425665855408,
"eval_mean_accuracy": 0.9669106724628387,
"eval_mean_iou": 0.48345533623141934,
"eval_overall_accuracy": 0.9669106724628387,
"eval_runtime": 2.4464,
"eval_samples_per_second": 30.657,
"eval_steps_per_second": 4.088,
"step": 1972
},
{
"epoch": 68.96551724137932,
"grad_norm": 2.6724536418914795,
"learning_rate": 1e-07,
"loss": 0.0826,
"step": 2000
},
{
"epoch": 68.96551724137932,
"eval_accuracy_no_text": 0.9701784379830324,
"eval_accuracy_text": NaN,
"eval_iou_no_text": 0.9701784379830324,
"eval_iou_text": 0.0,
"eval_loss": 0.12989187240600586,
"eval_mean_accuracy": 0.9701784379830324,
"eval_mean_iou": 0.4850892189915162,
"eval_overall_accuracy": 0.9701784379830324,
"eval_runtime": 2.4612,
"eval_samples_per_second": 30.473,
"eval_steps_per_second": 4.063,
"step": 2000
},
{
"epoch": 68.96551724137932,
"step": 2000,
"total_flos": 2.7210398151632486e+17,
"train_loss": 0.11668173718452454,
"train_runtime": 724.9682,
"train_samples_per_second": 22.07,
"train_steps_per_second": 2.759
}
],
"logging_steps": 100,
"max_steps": 2000,
"num_input_tokens_seen": 0,
"num_train_epochs": 69,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2.7210398151632486e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}