trainer_state.json

{
  "best_metric": 0.9024,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0604/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 6.735155628520579e-05,
      "loss": 1.8247,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.784,
      "eval_loss": 0.7968631386756897,
      "eval_runtime": 6.5834,
      "eval_samples_per_second": 569.611,
      "eval_steps_per_second": 8.962,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.4384562373161316,
      "learning_rate": 5.9777905730703695e-05,
      "loss": 0.5776,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8365333333333334,
      "eval_loss": 0.5670688152313232,
      "eval_runtime": 7.0287,
      "eval_samples_per_second": 533.527,
      "eval_steps_per_second": 8.394,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 4.8470148469326895e-05,
      "loss": 0.3149,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8626666666666667,
      "eval_loss": 0.4686274826526642,
      "eval_runtime": 6.7904,
      "eval_samples_per_second": 552.254,
      "eval_steps_per_second": 8.689,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 3.512382377646818e-05,
      "loss": 0.1753,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.4274883568286896,
      "eval_runtime": 6.581,
      "eval_samples_per_second": 569.818,
      "eval_steps_per_second": 8.965,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 3.9416255950927734,
      "learning_rate": 2.1720561940597914e-05,
      "loss": 0.0894,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.40647050738334656,
      "eval_runtime": 6.5734,
      "eval_samples_per_second": 570.478,
      "eval_steps_per_second": 8.976,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.060098472982645035,
      "learning_rate": 1.0338974170865825e-05,
      "loss": 0.0391,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.37632134556770325,
      "eval_runtime": 6.6315,
      "eval_samples_per_second": 565.48,
      "eval_steps_per_second": 8.897,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.6252084970474243,
      "learning_rate": 2.7118040308635984e-06,
      "loss": 0.0184,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9008,
      "eval_loss": 0.3739790618419647,
      "eval_runtime": 6.6473,
      "eval_samples_per_second": 564.135,
      "eval_steps_per_second": 8.876,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.04893786087632179,
      "learning_rate": 2.1903393706834692e-10,
      "loss": 0.0126,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9024,
      "eval_loss": 0.3709087371826172,
      "eval_runtime": 6.7114,
      "eval_samples_per_second": 558.753,
      "eval_steps_per_second": 8.791,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}