last-checkpoint/trainer_state.json

{
  "best_metric": 0.3712866943650337,
  "best_model_checkpoint": "/home2/s5432073/language-tech-project/results/ltp-roberta-large-default/checkpoint-1000",
  "epoch": 5.9171597633136095,
  "global_step": 1000,
  "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": 0.4444,
      "step": 200
    },
    {
      "epoch": 1.18,
      "eval_f1": 0.13967554851569386,
      "eval_f1_all": [
        0.1323529411764706,
        0.1882770870337478,
        0.0,
        0.0,
        0.49830890642615555,
        0.0,
        0.0,
        0.0,
        0.7215112736136504,
        0.6323024054982818,
        0.0,
        0.0043859649122807015,
        0.0,
        0.0,
        0.0,
        0.0,
        0.6163723916532904,
        0.0,
        0.0,
        0.0
      ],
      "eval_loss": 0.35773786902427673,
      "eval_runtime": 4.4833,
      "eval_samples_per_second": 422.907,
      "eval_steps_per_second": 13.383,
      "step": 200
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.037383177570094e-06,
      "loss": 0.3287,
      "step": 400
    },
    {
      "epoch": 2.37,
      "eval_f1": 0.26059605664667,
      "eval_f1_all": [
        0.4311377245508982,
        0.33333333333333337,
        0.0,
        0.0,
        0.6143001007049347,
        0.03592814371257485,
        0.2823529411764706,
        0.0,
        0.7357440890125174,
        0.55,
        0.1650485436893204,
        0.3709677419354839,
        0.0,
        0.0,
        0.11209439528023599,
        0.0,
        0.6120218579234973,
        0.5846153846153846,
        0.034934497816593885,
        0.3494423791821561
      ],
      "eval_loss": 0.32891085743904114,
      "eval_runtime": 3.0437,
      "eval_samples_per_second": 622.918,
      "eval_steps_per_second": 19.713,
      "step": 400
    },
    {
      "epoch": 3.55,
      "learning_rate": 6.791277258566978e-06,
      "loss": 0.2952,
      "step": 600
    },
    {
      "epoch": 3.55,
      "eval_f1": 0.32608326603750615,
      "eval_f1_all": [
        0.5166240409207161,
        0.5852478839177752,
        0.0,
        0.0,
        0.6099585062240663,
        0.0588235294117647,
        0.29069767441860467,
        0.0,
        0.745417515274949,
        0.619047619047619,
        0.3628691983122363,
        0.4564564564564564,
        0.0,
        0.015625,
        0.49217935349322217,
        0.04332129963898917,
        0.6925515055467513,
        0.6268656716417911,
        0.05714285714285715,
        0.3488372093023256
      ],
      "eval_loss": 0.31139805912971497,
      "eval_runtime": 2.9889,
      "eval_samples_per_second": 634.353,
      "eval_steps_per_second": 20.074,
      "step": 600
    },
    {
      "epoch": 4.73,
      "learning_rate": 5.545171339563863e-06,
      "loss": 0.2745,
      "step": 800
    },
    {
      "epoch": 4.73,
      "eval_f1": 0.35306069701657317,
      "eval_f1_all": [
        0.5316455696202532,
        0.575,
        0.02857142857142857,
        0.125,
        0.6256306760847629,
        0.1005586592178771,
        0.32222222222222224,
        0.0,
        0.7624595469255664,
        0.6393442622950821,
        0.4034334763948497,
        0.5371900826446281,
        0.0,
        0.015503875968992248,
        0.5170630816959669,
        0.08304498269896193,
        0.6463620981387478,
        0.6822429906542056,
        0.07407407407407407,
        0.39186691312384475
      ],
      "eval_loss": 0.3067249059677124,
      "eval_runtime": 2.9637,
      "eval_samples_per_second": 639.733,
      "eval_steps_per_second": 20.245,
      "step": 800
    },
    {
      "epoch": 5.92,
      "learning_rate": 4.299065420560748e-06,
      "loss": 0.2575,
      "step": 1000
    },
    {
      "epoch": 5.92,
      "eval_f1": 0.3712866943650337,
      "eval_f1_all": [
        0.5213032581453635,
        0.5556978233034571,
        0.1081081081081081,
        0.1391304347826087,
        0.6363636363636365,
        0.1005586592178771,
        0.2840909090909091,
        0.0,
        0.7576530612244897,
        0.6426076833527357,
        0.4453441295546559,
        0.5411140583554377,
        0.0,
        0.015037593984962405,
        0.5839929639401935,
        0.1921921921921922,
        0.641573994867408,
        0.6936936936936936,
        0.12648221343873517,
        0.4407894736842105
      ],
      "eval_loss": 0.3082274794578552,
      "eval_runtime": 2.9599,
      "eval_samples_per_second": 640.56,
      "eval_steps_per_second": 20.271,
      "step": 1000
    }
  ],
  "max_steps": 1690,
  "num_train_epochs": 10,
  "total_flos": 5502978001057296.0,
  "trial_name": null,
  "trial_params": null
}