trainer_state.json

{
  "best_metric": 0.7441860465116279,
  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-6",
  "epoch": 9.6,
  "eval_steps": 500,
  "global_step": 60,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.96,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5859283208847046,
      "eval_runtime": 0.2566,
      "eval_samples_per_second": 167.599,
      "eval_steps_per_second": 42.874,
      "step": 6
    },
    {
      "epoch": 1.6,
      "grad_norm": 3.7175180912017822,
      "learning_rate": 4.62962962962963e-06,
      "loss": 0.605,
      "step": 10
    },
    {
      "epoch": 1.92,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5841977000236511,
      "eval_runtime": 0.2709,
      "eval_samples_per_second": 158.721,
      "eval_steps_per_second": 40.603,
      "step": 12
    },
    {
      "epoch": 2.88,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5918598771095276,
      "eval_runtime": 0.2627,
      "eval_samples_per_second": 163.663,
      "eval_steps_per_second": 41.867,
      "step": 18
    },
    {
      "epoch": 3.2,
      "grad_norm": 2.415996551513672,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.5428,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5884882807731628,
      "eval_runtime": 0.9618,
      "eval_samples_per_second": 44.706,
      "eval_steps_per_second": 11.436,
      "step": 25
    },
    {
      "epoch": 4.8,
      "grad_norm": 2.7047529220581055,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.5584,
      "step": 30
    },
    {
      "epoch": 4.96,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.588610827922821,
      "eval_runtime": 0.2732,
      "eval_samples_per_second": 157.373,
      "eval_steps_per_second": 40.258,
      "step": 31
    },
    {
      "epoch": 5.92,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5914657711982727,
      "eval_runtime": 0.2503,
      "eval_samples_per_second": 171.764,
      "eval_steps_per_second": 43.94,
      "step": 37
    },
    {
      "epoch": 6.4,
      "grad_norm": 3.9282095432281494,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.5593,
      "step": 40
    },
    {
      "epoch": 6.88,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5934743881225586,
      "eval_runtime": 0.2654,
      "eval_samples_per_second": 162.014,
      "eval_steps_per_second": 41.446,
      "step": 43
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.82995343208313,
      "learning_rate": 9.259259259259259e-07,
      "loss": 0.5097,
      "step": 50
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5947032570838928,
      "eval_runtime": 0.2582,
      "eval_samples_per_second": 166.523,
      "eval_steps_per_second": 42.599,
      "step": 50
    },
    {
      "epoch": 8.96,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.594927966594696,
      "eval_runtime": 0.2847,
      "eval_samples_per_second": 151.012,
      "eval_steps_per_second": 38.631,
      "step": 56
    },
    {
      "epoch": 9.6,
      "grad_norm": 3.3591854572296143,
      "learning_rate": 0.0,
      "loss": 0.5205,
      "step": 60
    },
    {
      "epoch": 9.6,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.5949187874794006,
      "eval_runtime": 0.2565,
      "eval_samples_per_second": 167.627,
      "eval_steps_per_second": 42.881,
      "step": 60
    },
    {
      "epoch": 9.6,
      "step": 60,
      "total_flos": 7.3694882123477e+16,
      "train_loss": 0.5492916504542033,
      "train_runtime": 34.9653,
      "train_samples_per_second": 28.314,
      "train_steps_per_second": 1.716
    }
  ],
  "logging_steps": 10,
  "max_steps": 60,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3694882123477e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}