last-checkpoint/trainer_state.json

{
  "best_metric": 0.3950704777630208,
  "best_model_checkpoint": "/home2/s5432073/language-tech-project/results/ltp-roberta-large-defaultltp-roberta-large-default-9/checkpoint-1200",
  "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": 0.4412,
      "step": 200
    },
    {
      "epoch": 1.18,
      "eval_f1": 0.15828796157072217,
      "eval_f1_all": [
        0.2097902097902098,
        0.2789115646258503,
        0.0,
        0.0,
        0.547798066595059,
        0.0,
        0.0,
        0.0,
        0.7165913492575855,
        0.636150234741784,
        0.0,
        0.013071895424836602,
        0.0,
        0.0,
        0.0,
        0.0,
        0.6257166257166257,
        0.13235294117647056,
        0.0,
        0.005376344086021506
      ],
      "eval_loss": 0.35295796394348145,
      "eval_runtime": 2.6168,
      "eval_samples_per_second": 724.543,
      "eval_steps_per_second": 22.929,
      "step": 200
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.037383177570094e-06,
      "loss": 0.3276,
      "step": 400
    },
    {
      "epoch": 2.37,
      "eval_f1": 0.27070745440431776,
      "eval_f1_all": [
        0.3987730061349693,
        0.37480314960629924,
        0.0,
        0.0,
        0.6236125126135217,
        0.05917159763313609,
        0.3404255319148936,
        0.0,
        0.7367021276595745,
        0.5967130214917826,
        0.11518324607329843,
        0.4112149532710281,
        0.0,
        0.0,
        0.15251798561151078,
        0.0,
        0.6607431340872374,
        0.624390243902439,
        0.017621145374449337,
        0.30227743271221535
      ],
      "eval_loss": 0.3264588415622711,
      "eval_runtime": 2.8315,
      "eval_samples_per_second": 669.617,
      "eval_steps_per_second": 21.19,
      "step": 400
    },
    {
      "epoch": 3.55,
      "learning_rate": 6.791277258566978e-06,
      "loss": 0.2941,
      "step": 600
    },
    {
      "epoch": 3.55,
      "eval_f1": 0.3284303777422183,
      "eval_f1_all": [
        0.5230769230769231,
        0.5660847880299252,
        0.014388489208633093,
        0.09090909090909091,
        0.6258776328986961,
        0.0588235294117647,
        0.2923976608187135,
        0.0,
        0.7289595758780648,
        0.6356589147286822,
        0.28971962616822433,
        0.40816326530612246,
        0.0,
        0.015625,
        0.47457627118644063,
        0.06382978723404255,
        0.6955153422501967,
        0.635897435897436,
        0.09795918367346938,
        0.3511450381679389
      ],
      "eval_loss": 0.31264743208885193,
      "eval_runtime": 2.8103,
      "eval_samples_per_second": 674.672,
      "eval_steps_per_second": 21.35,
      "step": 600
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.545171339563863e-06,
      "loss": 0.2732,
      "step": 800
    },
    {
      "epoch": 4.73,
      "eval_f1": 0.3619965683236828,
      "eval_f1_all": [
        0.5528089887640449,
        0.5821596244131456,
        0.0945945945945946,
        0.18965517241379312,
        0.6472868217054263,
        0.13829787234042554,
        0.3692307692307692,
        0.0,
        0.7455386649041639,
        0.638095238095238,
        0.37606837606837606,
        0.4579025110782866,
        0.0,
        0.015503875968992248,
        0.5087179487179486,
        0.07719298245614035,
        0.6716917922948074,
        0.6666666666666667,
        0.08298755186721991,
        0.425531914893617
      ],
      "eval_loss": 0.3080659508705139,
      "eval_runtime": 2.7772,
      "eval_samples_per_second": 682.698,
      "eval_steps_per_second": 21.604,
      "step": 800
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.299065420560748e-06,
      "loss": 0.2559,
      "step": 1000
    },
    {
      "epoch": 5.92,
      "eval_f1": 0.37526293741944416,
      "eval_f1_all": [
        0.5243619489559165,
        0.5671641791044777,
        0.06896551724137931,
        0.1724137931034483,
        0.6521327014218009,
        0.11363636363636363,
        0.3711340206185567,
        0.0,
        0.7530312699425653,
        0.6465116279069768,
        0.42857142857142855,
        0.5034387895460799,
        0.0,
        0.015384615384615385,
        0.5711743772241993,
        0.17629179331306988,
        0.6535162950257289,
        0.6972477064220184,
        0.11290322580645162,
        0.4773790951638066
      ],
      "eval_loss": 0.30820420384407043,
      "eval_runtime": 2.6936,
      "eval_samples_per_second": 703.901,
      "eval_steps_per_second": 22.275,
      "step": 1000
    },
    {
      "epoch": 7.1,
      "learning_rate": 3.0529595015576325e-06,
      "loss": 0.2414,
      "step": 1200
    },
    {
      "epoch": 7.1,
      "eval_f1": 0.3950704777630208,
      "eval_f1_all": [
        0.5097087378640777,
        0.5710872162485066,
        0.13924050632911392,
        0.25757575757575757,
        0.6406406406406406,
        0.17346938775510204,
        0.3553299492385787,
        0.0,
        0.7513020833333334,
        0.6586586586586587,
        0.47727272727272735,
        0.5625,
        0.0,
        0.015267175572519085,
        0.5554520037278659,
        0.24444444444444446,
        0.6672012830793906,
        0.7079646017699114,
        0.14925373134328354,
        0.4650406504065041
      ],
      "eval_loss": 0.309607595205307,
      "eval_runtime": 2.8834,
      "eval_samples_per_second": 657.552,
      "eval_steps_per_second": 20.809,
      "step": 1200
    },
    {
      "epoch": 8.28,
      "learning_rate": 1.8068535825545173e-06,
      "loss": 0.2305,
      "step": 1400
    },
    {
      "epoch": 8.28,
      "eval_f1": 0.38770117434745416,
      "eval_f1_all": [
        0.5200945626477541,
        0.5766590389016018,
        0.11842105263157895,
        0.22399999999999998,
        0.6587771203155818,
        0.1658031088082902,
        0.380952380952381,
        0.0,
        0.7496774193548387,
        0.6578657865786579,
        0.4302788844621514,
        0.5163043478260869,
        0.0,
        0.015267175572519085,
        0.5784832451499118,
        0.1567398119122257,
        0.7061435973353072,
        0.690909090909091,
        0.1450381679389313,
        0.4626086956521739
      ],
      "eval_loss": 0.3088609278202057,
      "eval_runtime": 2.6739,
      "eval_samples_per_second": 709.087,
      "eval_steps_per_second": 22.439,
      "step": 1400
    },
    {
      "epoch": 9.47,
      "learning_rate": 5.607476635514019e-07,
      "loss": 0.2244,
      "step": 1600
    },
    {
      "epoch": 9.47,
      "eval_f1": 0.39269551799562785,
      "eval_f1_all": [
        0.5283018867924528,
        0.5794392523364486,
        0.11612903225806452,
        0.22047244094488186,
        0.6502369668246445,
        0.19095477386934676,
        0.3864734299516908,
        0.0,
        0.7457404980340759,
        0.6617179215270413,
        0.43307086614173235,
        0.5231607629427792,
        0.0,
        0.015151515151515152,
        0.5685740236148956,
        0.2138728323699422,
        0.6947852760736196,
        0.7031963470319635,
        0.14869888475836432,
        0.47393364928909953
      ],
      "eval_loss": 0.3094000220298767,
      "eval_runtime": 3.0101,
      "eval_samples_per_second": 629.877,
      "eval_steps_per_second": 19.933,
      "step": 1600
    }
  ],
  "max_steps": 1690,
  "num_train_epochs": 10,
  "total_flos": 8787505912201752.0,
  "trial_name": null,
  "trial_params": null
}