trainer_state.json

{
  "best_metric": 0.932,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0990/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": 0.28995150327682495,
      "learning_rate": 7e-05,
      "loss": 0.7274,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9114666666666666,
      "eval_loss": 0.29076045751571655,
      "eval_runtime": 14.147,
      "eval_samples_per_second": 265.073,
      "eval_steps_per_second": 4.17,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3359301090240479,
      "learning_rate": 7e-05,
      "loss": 0.1575,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9237333333333333,
      "eval_loss": 0.25154751539230347,
      "eval_runtime": 13.4355,
      "eval_samples_per_second": 279.111,
      "eval_steps_per_second": 4.391,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.5521812438964844,
      "learning_rate": 7e-05,
      "loss": 0.0886,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9314666666666667,
      "eval_loss": 0.23687158524990082,
      "eval_runtime": 13.4601,
      "eval_samples_per_second": 278.602,
      "eval_steps_per_second": 4.383,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.0633,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9288,
      "eval_loss": 0.26691338419914246,
      "eval_runtime": 13.4912,
      "eval_samples_per_second": 277.96,
      "eval_steps_per_second": 4.373,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.29588252305984497,
      "learning_rate": 7e-05,
      "loss": 0.0404,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9192,
      "eval_loss": 0.3257554769515991,
      "eval_runtime": 13.3529,
      "eval_samples_per_second": 280.839,
      "eval_steps_per_second": 4.419,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.010991635732352734,
      "learning_rate": 7e-05,
      "loss": 0.0344,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.932,
      "eval_loss": 0.29125866293907166,
      "eval_runtime": 13.9478,
      "eval_samples_per_second": 268.859,
      "eval_steps_per_second": 4.23,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}