trainer_state.json

{
  "best_metric": 0.8808,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0700/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 2.266,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7544,
      "eval_loss": 1.077734351158142,
      "eval_runtime": 6.1835,
      "eval_samples_per_second": 606.456,
      "eval_steps_per_second": 9.542,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.7756,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8314666666666667,
      "eval_loss": 0.6424756646156311,
      "eval_runtime": 6.711,
      "eval_samples_per_second": 558.782,
      "eval_steps_per_second": 8.791,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.830326557159424,
      "learning_rate": 3e-05,
      "loss": 0.4671,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.5112996697425842,
      "eval_runtime": 6.4041,
      "eval_samples_per_second": 585.561,
      "eval_steps_per_second": 9.213,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 50.036685943603516,
      "learning_rate": 3e-05,
      "loss": 0.3294,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8632,
      "eval_loss": 0.47524285316467285,
      "eval_runtime": 6.2346,
      "eval_samples_per_second": 601.486,
      "eval_steps_per_second": 9.463,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.236,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.42880597710609436,
      "eval_runtime": 6.0812,
      "eval_samples_per_second": 616.658,
      "eval_steps_per_second": 9.702,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.310635566711426,
      "learning_rate": 3e-05,
      "loss": 0.1782,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8808,
      "eval_loss": 0.4088763892650604,
      "eval_runtime": 6.5025,
      "eval_samples_per_second": 576.704,
      "eval_steps_per_second": 9.073,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}