trainer_state.json

{
  "best_metric": 0.813953488372093,
  "best_model_checkpoint": "vit-base-patch16-224/checkpoint-12",
  "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": 0.7906976744186046,
      "eval_loss": 0.4999209940433502,
      "eval_runtime": 0.2683,
      "eval_samples_per_second": 160.24,
      "eval_steps_per_second": 40.992,
      "step": 6
    },
    {
      "epoch": 1.6,
      "grad_norm": 3.087608575820923,
      "learning_rate": 4.62962962962963e-06,
      "loss": 0.5993,
      "step": 10
    },
    {
      "epoch": 1.92,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4812479317188263,
      "eval_runtime": 0.3038,
      "eval_samples_per_second": 141.519,
      "eval_steps_per_second": 36.202,
      "step": 12
    },
    {
      "epoch": 2.88,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.48915746808052063,
      "eval_runtime": 0.2696,
      "eval_samples_per_second": 159.518,
      "eval_steps_per_second": 40.807,
      "step": 18
    },
    {
      "epoch": 3.2,
      "grad_norm": 5.1914448738098145,
      "learning_rate": 3.7037037037037037e-06,
      "loss": 0.6031,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.49750375747680664,
      "eval_runtime": 0.2506,
      "eval_samples_per_second": 171.576,
      "eval_steps_per_second": 43.892,
      "step": 25
    },
    {
      "epoch": 4.8,
      "grad_norm": 4.207640171051025,
      "learning_rate": 2.7777777777777783e-06,
      "loss": 0.5946,
      "step": 30
    },
    {
      "epoch": 4.96,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5062411427497864,
      "eval_runtime": 0.263,
      "eval_samples_per_second": 163.481,
      "eval_steps_per_second": 41.821,
      "step": 31
    },
    {
      "epoch": 5.92,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5119370222091675,
      "eval_runtime": 0.2714,
      "eval_samples_per_second": 158.421,
      "eval_steps_per_second": 40.526,
      "step": 37
    },
    {
      "epoch": 6.4,
      "grad_norm": 4.061374664306641,
      "learning_rate": 1.8518518518518519e-06,
      "loss": 0.5557,
      "step": 40
    },
    {
      "epoch": 6.88,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5102289915084839,
      "eval_runtime": 0.2732,
      "eval_samples_per_second": 157.394,
      "eval_steps_per_second": 40.264,
      "step": 43
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.359290361404419,
      "learning_rate": 9.259259259259259e-07,
      "loss": 0.583,
      "step": 50
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5075170993804932,
      "eval_runtime": 0.2683,
      "eval_samples_per_second": 160.255,
      "eval_steps_per_second": 40.996,
      "step": 50
    },
    {
      "epoch": 8.96,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5080327987670898,
      "eval_runtime": 0.2647,
      "eval_samples_per_second": 162.473,
      "eval_steps_per_second": 41.563,
      "step": 56
    },
    {
      "epoch": 9.6,
      "grad_norm": 4.910851001739502,
      "learning_rate": 0.0,
      "loss": 0.6013,
      "step": 60
    },
    {
      "epoch": 9.6,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5078577399253845,
      "eval_runtime": 0.2596,
      "eval_samples_per_second": 165.63,
      "eval_steps_per_second": 42.371,
      "step": 60
    },
    {
      "epoch": 9.6,
      "step": 60,
      "total_flos": 7.3694882123477e+16,
      "train_loss": 0.5894826014836629,
      "train_runtime": 33.2904,
      "train_samples_per_second": 29.738,
      "train_steps_per_second": 1.802
    }
  ],
  "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
}