last-checkpoint/trainer_state.json

{
  "best_metric": 0.40153906255014676,
  "best_model_checkpoint": "/home2/s5432073/language-tech-project/results/ltp-roberta-large-defaultltp-roberta-large-default-0/checkpoint-1200",
  "epoch": 8.284023668639053,
  "global_step": 1400,
  "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": 0.4442,
      "step": 200
    },
    {
      "epoch": 1.18,
      "eval_f1": 0.14004634931160595,
      "eval_f1_all": [
        0.11111111111111112,
        0.19823008849557522,
        0.0,
        0.0,
        0.5077262693156732,
        0.0,
        0.015037593984962407,
        0.0,
        0.7197452229299364,
        0.6306954436450838,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.0,
        0.5973799126637556,
        0.015625,
        0.0,
        0.005376344086021506
      ],
      "eval_loss": 0.3573361933231354,
      "eval_runtime": 3.7892,
      "eval_samples_per_second": 500.363,
      "eval_steps_per_second": 15.834,
      "step": 200
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.037383177570094e-06,
      "loss": 0.3303,
      "step": 400
    },
    {
      "epoch": 2.37,
      "eval_f1": 0.26419617591690636,
      "eval_f1_all": [
        0.43804034582132567,
        0.31333333333333335,
        0.0,
        0.0,
        0.616,
        0.03592814371257485,
        0.3128491620111732,
        0.0,
        0.7331058020477816,
        0.5398601398601399,
        0.16580310880829016,
        0.391025641025641,
        0.0,
        0.0,
        0.13119533527696794,
        0.0,
        0.6373056994818653,
        0.6169154228855722,
        0.02643171806167401,
        0.32612966601178783
      ],
      "eval_loss": 0.32742321491241455,
      "eval_runtime": 3.2193,
      "eval_samples_per_second": 588.954,
      "eval_steps_per_second": 18.638,
      "step": 400
    },
    {
      "epoch": 3.55,
      "learning_rate": 6.791277258566978e-06,
      "loss": 0.2955,
      "step": 600
    },
    {
      "epoch": 3.55,
      "eval_f1": 0.33856925982901875,
      "eval_f1_all": [
        0.5025125628140703,
        0.5899632802937577,
        0.0,
        0.07407407407407407,
        0.6084275436793423,
        0.0588235294117647,
        0.3567567567567568,
        0.0,
        0.7440944881889764,
        0.6120481927710844,
        0.3916666666666666,
        0.49431818181818177,
        0.0,
        0.015625,
        0.5071868583162218,
        0.06382978723404255,
        0.6857597454256166,
        0.6403940886699507,
        0.10526315789473682,
        0.3206412825651303
      ],
      "eval_loss": 0.31045234203338623,
      "eval_runtime": 3.1516,
      "eval_samples_per_second": 601.599,
      "eval_steps_per_second": 19.038,
      "step": 600
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.545171339563863e-06,
      "loss": 0.275,
      "step": 800
    },
    {
      "epoch": 4.73,
      "eval_f1": 0.35946421270983664,
      "eval_f1_all": [
        0.5432692307692307,
        0.5728770595690748,
        0.055944055944055944,
        0.125,
        0.6434782608695652,
        0.11891891891891894,
        0.33513513513513515,
        0.0,
        0.7636594663278272,
        0.638655462184874,
        0.41071428571428564,
        0.5251396648044693,
        0.0,
        0.015503875968992248,
        0.5346733668341709,
        0.07017543859649122,
        0.657762938230384,
        0.6666666666666667,
        0.11244979919678717,
        0.39926062846580407
      ],
      "eval_loss": 0.3056282103061676,
      "eval_runtime": 3.1101,
      "eval_samples_per_second": 609.632,
      "eval_steps_per_second": 19.292,
      "step": 800
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.299065420560748e-06,
      "loss": 0.2582,
      "step": 1000
    },
    {
      "epoch": 5.92,
      "eval_f1": 0.3779236416443889,
      "eval_f1_all": [
        0.5119617224880383,
        0.5549872122762148,
        0.09523809523809525,
        0.24793388429752064,
        0.6457739791073124,
        0.10112359550561797,
        0.31351351351351353,
        0.0,
        0.7588785046728972,
        0.6312056737588652,
        0.44999999999999996,
        0.5300546448087431,
        0.0,
        0.015384615384615385,
        0.5868995633187772,
        0.17857142857142858,
        0.6494401378122309,
        0.7022222222222223,
        0.12851405622489961,
        0.45676998368678634
      ],
      "eval_loss": 0.30811235308647156,
      "eval_runtime": 3.5499,
      "eval_samples_per_second": 534.102,
      "eval_steps_per_second": 16.902,
      "step": 1000
    },
    {
      "epoch": 7.1,
      "learning_rate": 3.0529595015576325e-06,
      "loss": 0.2446,
      "step": 1200
    },
    {
      "epoch": 7.1,
      "eval_f1": 0.40153906255014676,
      "eval_f1_all": [
        0.5172413793103449,
        0.5871121718377088,
        0.12,
        0.3759398496240602,
        0.630952380952381,
        0.17258883248730966,
        0.3768115942028986,
        0.0,
        0.7563683866753755,
        0.6462513199577613,
        0.4758064516129032,
        0.5527369826435248,
        0.0,
        0.015037593984962405,
        0.5758683729433272,
        0.21203438395415472,
        0.6746794871794872,
        0.7136563876651982,
        0.17777777777777778,
        0.44991789819376027
      ],
      "eval_loss": 0.30632713437080383,
      "eval_runtime": 3.2144,
      "eval_samples_per_second": 589.85,
      "eval_steps_per_second": 18.666,
      "step": 1200
    },
    {
      "epoch": 8.28,
      "learning_rate": 1.8068535825545173e-06,
      "loss": 0.2344,
      "step": 1400
    },
    {
      "epoch": 8.28,
      "eval_f1": 0.39818355623683893,
      "eval_f1_all": [
        0.5059665871121718,
        0.6030267753201397,
        0.13071895424836602,
        0.3609022556390977,
        0.6425073457394711,
        0.17616580310880828,
        0.3827751196172249,
        0.0,
        0.7609942638623327,
        0.6363636363636364,
        0.4672131147540983,
        0.5517241379310345,
        0.0,
        0.015151515151515152,
        0.5907099035933392,
        0.14012738853503187,
        0.7006960556844547,
        0.7053571428571428,
        0.1484375,
        0.4448336252189142
      ],
      "eval_loss": 0.30558517575263977,
      "eval_runtime": 3.0733,
      "eval_samples_per_second": 616.921,
      "eval_steps_per_second": 19.523,
      "step": 1400
    }
  ],
  "max_steps": 1690,
  "num_train_epochs": 10,
  "total_flos": 7696217086325736.0,
  "trial_name": null,
  "trial_params": null
}