last-checkpoint/trainer_state.json

{
  "best_metric": 0.37634159533224454,
  "best_model_checkpoint": "/home2/s5432073/language-tech-project/results/ltp-roberta-large-defaultltp-roberta-large-default-2/checkpoint-1600",
  "epoch": 9.467455621301776,
  "global_step": 1600,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.18,
      "learning_rate": 9.28348909657321e-06,
      "loss": 0.4492,
      "step": 200
    },
    {
      "epoch": 1.18,
      "eval_f1": 0.11572875154130194,
      "eval_f1_all": [
        0.0,
        0.05847953216374269,
        0.0,
        0.0,
        0.4682713347921225,
        0.0,
        0.0,
        0.0,
        0.6747759282970549,
        0.520393811533052,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.5926544240400667,
        0.0,
        0.0,
        0.0
      ],
      "eval_loss": 0.36748331785202026,
      "eval_runtime": 3.8035,
      "eval_samples_per_second": 498.483,
      "eval_steps_per_second": 15.775,
      "step": 200
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.037383177570094e-06,
      "loss": 0.3449,
      "step": 400
    },
    {
      "epoch": 2.37,
      "eval_f1": 0.22644858426717676,
      "eval_f1_all": [
        0.3419354838709677,
        0.30434782608695654,
        0.0,
        0.0,
        0.5748898678414096,
        0.0,
        0.3260869565217391,
        0.0,
        0.7539203860072377,
        0.5495118549511855,
        0.0879120879120879,
        0.23971377459749552,
        0.0,
        0.0,
        0.055045871559633024,
        0.0,
        0.6266548984995588,
        0.5684210526315789,
        0.008928571428571428,
        0.09160305343511452
      ],
      "eval_loss": 0.33438944816589355,
      "eval_runtime": 2.9482,
      "eval_samples_per_second": 643.096,
      "eval_steps_per_second": 20.351,
      "step": 400
    },
    {
      "epoch": 3.55,
      "learning_rate": 6.791277258566978e-06,
      "loss": 0.3163,
      "step": 600
    },
    {
      "epoch": 3.55,
      "eval_f1": 0.29790838893606486,
      "eval_f1_all": [
        0.4838709677419356,
        0.4791965566714491,
        0.0,
        0.0,
        0.5882352941176471,
        0.012121212121212121,
        0.303030303030303,
        0.0,
        0.7144866385372716,
        0.574385510996119,
        0.3069767441860465,
        0.4,
        0.0,
        0.0,
        0.40190249702734837,
        0.0,
        0.6844840386043058,
        0.6761904761904762,
        0.034482758620689655,
        0.29880478087649404
      ],
      "eval_loss": 0.3200507164001465,
      "eval_runtime": 2.9119,
      "eval_samples_per_second": 651.111,
      "eval_steps_per_second": 20.605,
      "step": 600
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.545171339563863e-06,
      "loss": 0.3021,
      "step": 800
    },
    {
      "epoch": 4.73,
      "eval_f1": 0.33206438169614216,
      "eval_f1_all": [
        0.5162907268170427,
        0.5621761658031088,
        0.0,
        0.0,
        0.6198347107438016,
        0.08139534883720931,
        0.36734693877551017,
        0.0,
        0.7533632286995516,
        0.6167290886392011,
        0.36111111111111105,
        0.42990654205607476,
        0.0,
        0.0,
        0.5192909280500522,
        0.014814814814814815,
        0.6877828054298644,
        0.6568627450980392,
        0.08906882591093117,
        0.36531365313653136
      ],
      "eval_loss": 0.3112545609474182,
      "eval_runtime": 3.7309,
      "eval_samples_per_second": 508.183,
      "eval_steps_per_second": 16.082,
      "step": 800
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.299065420560748e-06,
      "loss": 0.2879,
      "step": 1000
    },
    {
      "epoch": 5.92,
      "eval_f1": 0.3556641957175606,
      "eval_f1_all": [
        0.5228426395939088,
        0.5761843790012804,
        0.014285714285714285,
        0.05660377358490566,
        0.6353166986564299,
        0.09195402298850576,
        0.4059405940594059,
        0.0,
        0.7578814627994955,
        0.639225181598063,
        0.3873873873873874,
        0.49717514124293793,
        0.0,
        0.015503875968992248,
        0.5467059980334316,
        0.08934707903780069,
        0.6688907422852377,
        0.6995515695067265,
        0.09876543209876544,
        0.4097222222222222
      ],
      "eval_loss": 0.3077404499053955,
      "eval_runtime": 3.283,
      "eval_samples_per_second": 577.521,
      "eval_steps_per_second": 18.276,
      "step": 1000
    },
    {
      "epoch": 7.1,
      "learning_rate": 3.0529595015576325e-06,
      "loss": 0.2783,
      "step": 1200
    },
    {
      "epoch": 7.1,
      "eval_f1": 0.3662892905615983,
      "eval_f1_all": [
        0.5064599483204134,
        0.5771643663739021,
        0.028169014084507046,
        0.125,
        0.6408730158730159,
        0.1222222222222222,
        0.35789473684210527,
        0.0,
        0.7578947368421052,
        0.6680988184747583,
        0.4291497975708502,
        0.5216178521617852,
        0.0,
        0.015503875968992248,
        0.5635864592863677,
        0.10135135135135136,
        0.6790622473726757,
        0.7058823529411764,
        0.10317460317460318,
        0.422680412371134
      ],
      "eval_loss": 0.3067641854286194,
      "eval_runtime": 3.0738,
      "eval_samples_per_second": 616.829,
      "eval_steps_per_second": 19.52,
      "step": 1200
    },
    {
      "epoch": 8.28,
      "learning_rate": 1.8068535825545173e-06,
      "loss": 0.2721,
      "step": 1400
    },
    {
      "epoch": 8.28,
      "eval_f1": 0.36379633008264506,
      "eval_f1_all": [
        0.5247524752475247,
        0.5838509316770186,
        0.028368794326241138,
        0.07407407407407407,
        0.6463654223968567,
        0.13259668508287292,
        0.35106382978723405,
        0.0,
        0.7543075941289088,
        0.6308243727598566,
        0.4,
        0.5212620027434842,
        0.0,
        0.015503875968992248,
        0.5748709122203098,
        0.07612456747404846,
        0.6950959488272921,
        0.6986899563318777,
        0.14785992217898833,
        0.42031523642732044
      ],
      "eval_loss": 0.3058605194091797,
      "eval_runtime": 3.1337,
      "eval_samples_per_second": 605.037,
      "eval_steps_per_second": 19.147,
      "step": 1400
    },
    {
      "epoch": 9.47,
      "learning_rate": 5.607476635514019e-07,
      "loss": 0.2672,
      "step": 1600
    },
    {
      "epoch": 9.47,
      "eval_f1": 0.37634159533224454,
      "eval_f1_all": [
        0.5393794749403341,
        0.5859085290482077,
        0.06896551724137931,
        0.15384615384615385,
        0.6564885496183206,
        0.12972972972972974,
        0.38000000000000006,
        0.0,
        0.7552715654952076,
        0.6600441501103753,
        0.41463414634146345,
        0.5263157894736842,
        0.0,
        0.015503875968992248,
        0.5658263305322129,
        0.10596026490066227,
        0.6955848179705655,
        0.7085201793721972,
        0.13076923076923078,
        0.4340836012861736
      ],
      "eval_loss": 0.3051939308643341,
      "eval_runtime": 3.2008,
      "eval_samples_per_second": 592.358,
      "eval_steps_per_second": 18.746,
      "step": 1600
    }
  ],
  "max_steps": 1690,
  "num_train_epochs": 10,
  "total_flos": 1.1949029604729552e+16,
  "trial_name": null,
  "trial_params": null
}