trainer_state.json

{
  "best_metric": 0.8669333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0372/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": 0.8008319139480591,
      "learning_rate": 0.0003,
      "loss": 0.6444,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.844,
      "eval_loss": 0.5228056907653809,
      "eval_runtime": 13.4936,
      "eval_samples_per_second": 277.909,
      "eval_steps_per_second": 4.372,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.3428,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.46705129742622375,
      "eval_runtime": 13.4377,
      "eval_samples_per_second": 279.065,
      "eval_steps_per_second": 4.391,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6319953203201294,
      "learning_rate": 0.0003,
      "loss": 0.2566,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8626666666666667,
      "eval_loss": 0.49877989292144775,
      "eval_runtime": 13.4516,
      "eval_samples_per_second": 278.778,
      "eval_steps_per_second": 4.386,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.08287152647972107,
      "learning_rate": 0.0003,
      "loss": 0.205,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8648,
      "eval_loss": 0.4688222408294678,
      "eval_runtime": 13.4869,
      "eval_samples_per_second": 278.048,
      "eval_steps_per_second": 4.375,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 31.411012649536133,
      "learning_rate": 0.0003,
      "loss": 0.1748,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.5320336222648621,
      "eval_runtime": 13.4781,
      "eval_samples_per_second": 278.229,
      "eval_steps_per_second": 4.377,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0370611846446991,
      "learning_rate": 0.0003,
      "loss": 0.1545,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8669333333333333,
      "eval_loss": 0.5038761496543884,
      "eval_runtime": 13.3964,
      "eval_samples_per_second": 279.926,
      "eval_steps_per_second": 4.404,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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
}