training_checkpoints/checkpoint-3215/trainer_state.json

{
  "best_metric": 0.7369315728111729,
  "best_model_checkpoint": "bart-base/checkpoint-2572",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 3215,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.78,
      "learning_rate": 4.871176775531364e-05,
      "loss": 1.2028,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7598760650658405,
      "eval_f1_macro": 0.6063644681337579,
      "eval_f1_micro": 0.7598760650658405,
      "eval_f1_weighted": 0.7489266744747745,
      "eval_loss": 0.8429557681083679,
      "eval_macro_fpr": 0.023226157303183535,
      "eval_macro_sensitivity": 0.6366904609623596,
      "eval_macro_specificity": 0.981694425945952,
      "eval_precision": 0.7600880594489849,
      "eval_precision_macro": 0.6003620126022547,
      "eval_recall": 0.7598760650658405,
      "eval_recall_macro": 0.6366904609623596,
      "eval_runtime": 43.5934,
      "eval_samples_per_second": 29.615,
      "eval_steps_per_second": 3.716,
      "eval_weighted_fpr": 0.02207348333808032,
      "eval_weighted_sensitivity": 0.7598760650658405,
      "eval_weighted_specificity": 0.9655403241234398,
      "step": 643
    },
    {
      "epoch": 1.56,
      "learning_rate": 4.741575946086055e-05,
      "loss": 0.715,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7931835786212239,
      "eval_f1_macro": 0.7213629796675549,
      "eval_f1_micro": 0.793183578621224,
      "eval_f1_weighted": 0.7853477061503886,
      "eval_loss": 0.7331996560096741,
      "eval_macro_fpr": 0.018982515099395962,
      "eval_macro_sensitivity": 0.7320748785168393,
      "eval_macro_specificity": 0.9845140063777027,
      "eval_precision": 0.8019650926098448,
      "eval_precision_macro": 0.7386000214414028,
      "eval_recall": 0.7931835786212239,
      "eval_recall_macro": 0.7320748785168393,
      "eval_runtime": 77.2393,
      "eval_samples_per_second": 16.714,
      "eval_steps_per_second": 2.097,
      "eval_weighted_fpr": 0.018283914264192288,
      "eval_weighted_sensitivity": 0.7931835786212239,
      "eval_weighted_specificity": 0.9745265170443169,
      "step": 1286
    },
    {
      "epoch": 2.33,
      "learning_rate": 4.611975116640747e-05,
      "loss": 0.578,
      "step": 1500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7939581719597212,
      "eval_f1_macro": 0.6998313050068915,
      "eval_f1_micro": 0.7939581719597211,
      "eval_f1_weighted": 0.79006368267302,
      "eval_loss": 0.8045273423194885,
      "eval_macro_fpr": 0.018545878882531782,
      "eval_macro_sensitivity": 0.7068834234702027,
      "eval_macro_specificity": 0.9847671550097226,
      "eval_precision": 0.8075005963568469,
      "eval_precision_macro": 0.7230632440844518,
      "eval_recall": 0.7939581719597212,
      "eval_recall_macro": 0.7068834234702027,
      "eval_runtime": 86.4099,
      "eval_samples_per_second": 14.94,
      "eval_steps_per_second": 1.875,
      "eval_weighted_fpr": 0.018199233716475097,
      "eval_weighted_sensitivity": 0.7939581719597212,
      "eval_weighted_specificity": 0.977549153186118,
      "step": 1929
    },
    {
      "epoch": 3.11,
      "learning_rate": 4.482633488854329e-05,
      "loss": 0.481,
      "step": 2000
    },
    {
      "epoch": 3.89,
      "learning_rate": 4.3530326594090203e-05,
      "loss": 0.3938,
      "step": 2500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8156467854376452,
      "eval_f1_macro": 0.7369315728111729,
      "eval_f1_micro": 0.8156467854376452,
      "eval_f1_weighted": 0.8105168035589256,
      "eval_loss": 0.8291025161743164,
      "eval_macro_fpr": 0.016938074954137138,
      "eval_macro_sensitivity": 0.7217605983180461,
      "eval_macro_specificity": 0.9857842035419169,
      "eval_precision": 0.8171220006778526,
      "eval_precision_macro": 0.7937034642937142,
      "eval_recall": 0.8156467854376452,
      "eval_recall_macro": 0.7217605983180461,
      "eval_runtime": 84.4366,
      "eval_samples_per_second": 15.29,
      "eval_steps_per_second": 1.919,
      "eval_weighted_fpr": 0.01588785046728972,
      "eval_weighted_sensitivity": 0.8156467854376452,
      "eval_weighted_specificity": 0.9711162676911109,
      "step": 2572
    },
    {
      "epoch": 4.67,
      "learning_rate": 4.2236910316226025e-05,
      "loss": 0.3238,
      "step": 3000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7939581719597212,
      "eval_f1_macro": 0.7360579650857649,
      "eval_f1_micro": 0.7939581719597211,
      "eval_f1_weighted": 0.7995200606905476,
      "eval_loss": 0.8888681530952454,
      "eval_macro_fpr": 0.018767728188278533,
      "eval_macro_sensitivity": 0.7515467286591376,
      "eval_macro_specificity": 0.9846803211687324,
      "eval_precision": 0.8146370555192632,
      "eval_precision_macro": 0.7464262358644205,
      "eval_recall": 0.7939581719597212,
      "eval_recall_macro": 0.7515467286591376,
      "eval_runtime": 82.179,
      "eval_samples_per_second": 15.71,
      "eval_steps_per_second": 1.971,
      "eval_weighted_fpr": 0.018199233716475097,
      "eval_weighted_sensitivity": 0.7939581719597212,
      "eval_weighted_specificity": 0.9762466455712672,
      "step": 3215
    }
  ],
  "logging_steps": 500,
  "max_steps": 19290,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 7879584476252160.0,
  "trial_name": null,
  "trial_params": null
}