trainer_state.json

{
  "best_metric": 0.76875,
  "best_model_checkpoint": "resnet-18-feature-extraction/checkpoint-16",
  "epoch": 7.8,
  "global_step": 16,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8,
      "eval_accuracy": 0.09375,
      "eval_f1": 0.0,
      "eval_loss": 1.4615901708602905,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc": 0.5,
      "eval_runtime": 4.0789,
      "eval_samples_per_second": 39.226,
      "eval_steps_per_second": 0.735,
      "step": 2
    },
    {
      "epoch": 1.8,
      "eval_accuracy": 0.09375,
      "eval_f1": 0.0,
      "eval_loss": 1.3914295434951782,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc": 0.5,
      "eval_runtime": 4.1829,
      "eval_samples_per_second": 38.251,
      "eval_steps_per_second": 0.717,
      "step": 4
    },
    {
      "epoch": 2.8,
      "eval_accuracy": 0.09375,
      "eval_f1": 0.0,
      "eval_loss": 1.473630666732788,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc": 0.5,
      "eval_runtime": 4.2309,
      "eval_samples_per_second": 37.817,
      "eval_steps_per_second": 0.709,
      "step": 6
    },
    {
      "epoch": 3.8,
      "eval_accuracy": 0.09375,
      "eval_f1": 0.0,
      "eval_loss": 1.6036533117294312,
      "eval_precision": 0.0,
      "eval_recall": 0.0,
      "eval_roc_auc": 0.5,
      "eval_runtime": 4.3657,
      "eval_samples_per_second": 36.65,
      "eval_steps_per_second": 0.687,
      "step": 8
    },
    {
      "epoch": 4.8,
      "learning_rate": 8.571428571428571e-06,
      "loss": 1.0775,
      "step": 10
    },
    {
      "epoch": 4.8,
      "eval_accuracy": 0.1375,
      "eval_f1": 0.09210526315789473,
      "eval_loss": 1.1970865726470947,
      "eval_precision": 1.0,
      "eval_recall": 0.04827586206896552,
      "eval_roc_auc": 0.5241379310344828,
      "eval_runtime": 4.1964,
      "eval_samples_per_second": 38.128,
      "eval_steps_per_second": 0.715,
      "step": 10
    },
    {
      "epoch": 5.8,
      "eval_accuracy": 0.28125,
      "eval_f1": 0.34285714285714286,
      "eval_loss": 0.9469502568244934,
      "eval_precision": 1.0,
      "eval_recall": 0.20689655172413793,
      "eval_roc_auc": 0.603448275862069,
      "eval_runtime": 4.128,
      "eval_samples_per_second": 38.76,
      "eval_steps_per_second": 0.727,
      "step": 12
    },
    {
      "epoch": 6.8,
      "eval_accuracy": 0.725,
      "eval_f1": 0.8211382113821138,
      "eval_loss": 0.6010352969169617,
      "eval_precision": 1.0,
      "eval_recall": 0.696551724137931,
      "eval_roc_auc": 0.8482758620689654,
      "eval_runtime": 4.0447,
      "eval_samples_per_second": 39.558,
      "eval_steps_per_second": 0.742,
      "step": 14
    },
    {
      "epoch": 7.8,
      "eval_accuracy": 0.76875,
      "eval_f1": 0.8537549407114624,
      "eval_loss": 0.566879153251648,
      "eval_precision": 1.0,
      "eval_recall": 0.7448275862068966,
      "eval_roc_auc": 0.8724137931034484,
      "eval_runtime": 4.1451,
      "eval_samples_per_second": 38.6,
      "eval_steps_per_second": 0.724,
      "step": 16
    },
    {
      "epoch": 7.8,
      "step": 16,
      "total_flos": 5.011254619486618e+16,
      "train_loss": 0.9726034700870514,
      "train_runtime": 60.475,
      "train_samples_per_second": 84.134,
      "train_steps_per_second": 0.265
    }
  ],
  "max_steps": 16,
  "num_train_epochs": 8,
  "total_flos": 5.011254619486618e+16,
  "trial_name": null,
  "trial_params": null
}