xlm-roberta-large-lemma-en / trainer_state.json
oliat's picture
Uploading model files
8ac591c verified
raw
history blame
8.02 kB
{
"best_metric": 0.9710087370929309,
"best_model_checkpoint": ".//debugged_en_ewt_ses_udpipe_16_0.1_0.00002_15_04-23-24_01-15/checkpoint-10803",
"epoch": 14.999398677089598,
"global_step": 12465,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 1.0,
"learning_rate": 1.985832550994282e-05,
"loss": 0.8748,
"step": 831
},
{
"epoch": 1.0,
"eval_accuracy": 0.9727829995913363,
"eval_f1": 0.9206887946094336,
"eval_loss": 0.11471811681985855,
"eval_precision": 0.9245188452285485,
"eval_recall": 0.9168903469529774,
"eval_runtime": 8.5451,
"eval_samples_per_second": 194.614,
"eval_steps_per_second": 24.341,
"step": 831
},
{
"epoch": 2.0,
"learning_rate": 1.8439873687804047e-05,
"loss": 0.074,
"step": 1662
},
{
"epoch": 2.0,
"eval_accuracy": 0.9862280343277483,
"eval_f1": 0.9563490092863883,
"eval_loss": 0.06378430128097534,
"eval_precision": 0.9554475094264735,
"eval_recall": 0.9572522119494979,
"eval_runtime": 9.0509,
"eval_samples_per_second": 183.738,
"eval_steps_per_second": 22.981,
"step": 1662
},
{
"epoch": 3.0,
"learning_rate": 1.7021421865665273e-05,
"loss": 0.0375,
"step": 2493
},
{
"epoch": 3.0,
"eval_accuracy": 0.9880261544748672,
"eval_f1": 0.961840405445692,
"eval_loss": 0.054762206971645355,
"eval_precision": 0.9614582298599385,
"eval_recall": 0.9622228849786261,
"eval_runtime": 8.6908,
"eval_samples_per_second": 191.352,
"eval_steps_per_second": 23.933,
"step": 2493
},
{
"epoch": 4.0,
"learning_rate": 1.5602970043526503e-05,
"loss": 0.0236,
"step": 3324
},
{
"epoch": 4.0,
"eval_accuracy": 0.9895790764201062,
"eval_f1": 0.9648930880159124,
"eval_loss": 0.054631613194942474,
"eval_precision": 0.9652770868570292,
"eval_recall": 0.964509394572025,
"eval_runtime": 8.7121,
"eval_samples_per_second": 190.884,
"eval_steps_per_second": 23.875,
"step": 3324
},
{
"epoch": 5.0,
"learning_rate": 1.4184518221387728e-05,
"loss": 0.0155,
"step": 4155
},
{
"epoch": 5.0,
"eval_accuracy": 0.989865140988966,
"eval_f1": 0.9674321916452189,
"eval_loss": 0.0665346086025238,
"eval_precision": 0.9633316904879251,
"eval_recall": 0.971567750273387,
"eval_runtime": 8.672,
"eval_samples_per_second": 191.767,
"eval_steps_per_second": 23.985,
"step": 4155
},
{
"epoch": 6.0,
"learning_rate": 1.2766066399248956e-05,
"loss": 0.0102,
"step": 4986
},
{
"epoch": 6.0,
"eval_accuracy": 0.989865140988966,
"eval_f1": 0.9660317302431989,
"eval_loss": 0.06522596627473831,
"eval_precision": 0.9665605095541401,
"eval_recall": 0.9655035291778507,
"eval_runtime": 9.1046,
"eval_samples_per_second": 182.656,
"eval_steps_per_second": 22.846,
"step": 4986
},
{
"epoch": 7.0,
"learning_rate": 1.1347614577110182e-05,
"loss": 0.0067,
"step": 5817
},
{
"epoch": 7.0,
"eval_accuracy": 0.9908868001634654,
"eval_f1": 0.9694357754990836,
"eval_loss": 0.06391981989145279,
"eval_precision": 0.9661334913112164,
"eval_recall": 0.9727607118003778,
"eval_runtime": 9.6275,
"eval_samples_per_second": 172.735,
"eval_steps_per_second": 21.605,
"step": 5817
},
{
"epoch": 8.0,
"learning_rate": 9.92916275497141e-06,
"loss": 0.005,
"step": 6648
},
{
"epoch": 8.0,
"eval_accuracy": 0.9906007355946056,
"eval_f1": 0.9695436507936508,
"eval_loss": 0.06800998747348785,
"eval_precision": 0.9675279675279675,
"eval_recall": 0.971567750273387,
"eval_runtime": 8.544,
"eval_samples_per_second": 194.64,
"eval_steps_per_second": 24.345,
"step": 6648
},
{
"epoch": 9.0,
"learning_rate": 8.510710932832637e-06,
"loss": 0.0033,
"step": 7479
},
{
"epoch": 9.0,
"eval_accuracy": 0.9906824683285655,
"eval_f1": 0.9697029899672196,
"eval_loss": 0.07487699389457703,
"eval_precision": 0.9689330024813896,
"eval_recall": 0.9704742022069788,
"eval_runtime": 8.5275,
"eval_samples_per_second": 195.016,
"eval_steps_per_second": 24.392,
"step": 7479
},
{
"epoch": 10.0,
"learning_rate": 7.092259110693864e-06,
"loss": 0.0027,
"step": 8310
},
{
"epoch": 10.0,
"eval_accuracy": 0.9906824683285655,
"eval_f1": 0.9683280381254964,
"eval_loss": 0.0723545178771019,
"eval_precision": 0.9670798215171046,
"eval_recall": 0.9695794810617357,
"eval_runtime": 8.98,
"eval_samples_per_second": 185.189,
"eval_steps_per_second": 23.163,
"step": 8310
},
{
"epoch": 11.0,
"learning_rate": 5.673807288555091e-06,
"loss": 0.0018,
"step": 9141
},
{
"epoch": 11.0,
"eval_accuracy": 0.9910911319983653,
"eval_f1": 0.969461852183428,
"eval_loss": 0.07421588897705078,
"eval_precision": 0.9700408082014532,
"eval_recall": 0.9688835868376579,
"eval_runtime": 9.1698,
"eval_samples_per_second": 181.357,
"eval_steps_per_second": 22.683,
"step": 9141
},
{
"epoch": 12.0,
"learning_rate": 4.255355466416318e-06,
"loss": 0.0018,
"step": 9972
},
{
"epoch": 12.0,
"eval_accuracy": 0.9909276665304454,
"eval_f1": 0.9694831013916502,
"eval_loss": 0.07263504713773727,
"eval_precision": 0.9693867408806282,
"eval_recall": 0.9695794810617357,
"eval_runtime": 8.7552,
"eval_samples_per_second": 189.945,
"eval_steps_per_second": 23.757,
"step": 9972
},
{
"epoch": 13.0,
"learning_rate": 2.8369036442775456e-06,
"loss": 0.001,
"step": 10803
},
{
"epoch": 13.0,
"eval_accuracy": 0.9913363302002453,
"eval_f1": 0.9710087370929309,
"eval_loss": 0.07499531656503677,
"eval_precision": 0.9697570649479424,
"eval_recall": 0.9722636444974649,
"eval_runtime": 8.5127,
"eval_samples_per_second": 195.356,
"eval_steps_per_second": 24.434,
"step": 10803
},
{
"epoch": 14.0,
"learning_rate": 1.4184518221387728e-06,
"loss": 0.0008,
"step": 11634
},
{
"epoch": 14.0,
"eval_accuracy": 0.9908868001634654,
"eval_f1": 0.969507350019865,
"eval_loss": 0.07384610921144485,
"eval_precision": 0.968641460752208,
"eval_recall": 0.9703747887463963,
"eval_runtime": 8.7778,
"eval_samples_per_second": 189.456,
"eval_steps_per_second": 23.696,
"step": 11634
},
{
"epoch": 15.0,
"learning_rate": 0.0,
"loss": 0.0007,
"step": 12465
},
{
"epoch": 15.0,
"eval_accuracy": 0.9910093992644053,
"eval_f1": 0.969751154820444,
"eval_loss": 0.07460830360651016,
"eval_precision": 0.9690291840381179,
"eval_recall": 0.9704742022069788,
"eval_runtime": 8.9902,
"eval_samples_per_second": 184.979,
"eval_steps_per_second": 23.136,
"step": 12465
},
{
"epoch": 15.0,
"step": 12465,
"total_flos": 1.722226508123076e+16,
"train_loss": 0.07061179929405056,
"train_runtime": 5194.8123,
"train_samples_per_second": 38.398,
"train_steps_per_second": 2.4
}
],
"max_steps": 12465,
"num_train_epochs": 15,
"total_flos": 1.722226508123076e+16,
"trial_name": null,
"trial_params": null
}