trainer_state.json

{
  "best_metric": 0.8994666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0353/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": 4.376876876876877e-05,
      "loss": 1.7138,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8162666666666667,
      "eval_loss": 0.7257247567176819,
      "eval_runtime": 5.2485,
      "eval_samples_per_second": 714.486,
      "eval_steps_per_second": 11.241,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 14.146262168884277,
      "learning_rate": 3.7518768768768766e-05,
      "loss": 0.461,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4689095616340637,
      "eval_runtime": 5.7707,
      "eval_samples_per_second": 649.84,
      "eval_steps_per_second": 10.224,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9600096344947815,
      "learning_rate": 3.126876876876877e-05,
      "loss": 0.2295,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.4555407464504242,
      "eval_runtime": 5.3157,
      "eval_samples_per_second": 705.456,
      "eval_steps_per_second": 11.099,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.7288385033607483,
      "learning_rate": 2.5018768768768767e-05,
      "loss": 0.1239,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.4010899066925049,
      "eval_runtime": 5.2923,
      "eval_samples_per_second": 708.581,
      "eval_steps_per_second": 11.148,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.1472112536430359,
      "learning_rate": 1.8768768768768768e-05,
      "loss": 0.0631,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8933333333333333,
      "eval_loss": 0.41405677795410156,
      "eval_runtime": 5.2185,
      "eval_samples_per_second": 718.595,
      "eval_steps_per_second": 11.306,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.016580581665039,
      "learning_rate": 1.2518768768768768e-05,
      "loss": 0.0299,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8994666666666666,
      "eval_loss": 0.40426769852638245,
      "eval_runtime": 5.2438,
      "eval_samples_per_second": 715.133,
      "eval_steps_per_second": 11.251,
      "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
}