trainer_state.json

{
  "best_metric": 0.38453333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0356/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 0.0003,
      "loss": 3.3178,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.22053333333333333,
      "eval_loss": 2.9166760444641113,
      "eval_runtime": 6.9317,
      "eval_samples_per_second": 540.991,
      "eval_steps_per_second": 8.512,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 20.597925186157227,
      "learning_rate": 0.0003,
      "loss": 2.7537,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.26826666666666665,
      "eval_loss": 2.71899676322937,
      "eval_runtime": 6.5399,
      "eval_samples_per_second": 573.402,
      "eval_steps_per_second": 9.022,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 25.493330001831055,
      "learning_rate": 0.0003,
      "loss": 2.5251,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3002666666666667,
      "eval_loss": 2.5705463886260986,
      "eval_runtime": 6.5159,
      "eval_samples_per_second": 575.518,
      "eval_steps_per_second": 9.055,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 22.36939811706543,
      "learning_rate": 0.0003,
      "loss": 2.346,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.33786666666666665,
      "eval_loss": 2.3929362297058105,
      "eval_runtime": 6.4048,
      "eval_samples_per_second": 585.497,
      "eval_steps_per_second": 9.212,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.73693084716797,
      "learning_rate": 0.0003,
      "loss": 2.2013,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.348,
      "eval_loss": 2.3033320903778076,
      "eval_runtime": 6.5462,
      "eval_samples_per_second": 572.849,
      "eval_steps_per_second": 9.013,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 27.72455406188965,
      "learning_rate": 0.0003,
      "loss": 2.0917,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3658666666666667,
      "eval_loss": 2.310325860977173,
      "eval_runtime": 6.9972,
      "eval_samples_per_second": 535.928,
      "eval_steps_per_second": 8.432,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 26.464313507080078,
      "learning_rate": 0.0003,
      "loss": 1.9582,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.38453333333333334,
      "eval_loss": 2.2385642528533936,
      "eval_runtime": 6.0606,
      "eval_samples_per_second": 618.748,
      "eval_steps_per_second": 9.735,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}