trainer_state.json

{
  "best_metric": 0.7890666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0260/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": 0.0003,
      "loss": 2.5423,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6034666666666667,
      "eval_loss": 1.4092572927474976,
      "eval_runtime": 12.5363,
      "eval_samples_per_second": 299.132,
      "eval_steps_per_second": 4.706,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 3.0307552814483643,
      "learning_rate": 0.0003,
      "loss": 1.1035,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6930666666666667,
      "eval_loss": 1.033342719078064,
      "eval_runtime": 11.4518,
      "eval_samples_per_second": 327.458,
      "eval_steps_per_second": 5.152,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.27029258012771606,
      "learning_rate": 0.0003,
      "loss": 0.7224,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7584,
      "eval_loss": 0.7946996092796326,
      "eval_runtime": 12.151,
      "eval_samples_per_second": 308.615,
      "eval_steps_per_second": 4.856,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.1108829975128174,
      "learning_rate": 0.0003,
      "loss": 0.5313,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7568,
      "eval_loss": 0.8136960864067078,
      "eval_runtime": 13.4619,
      "eval_samples_per_second": 278.565,
      "eval_steps_per_second": 4.383,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2450941801071167,
      "learning_rate": 0.0003,
      "loss": 0.4152,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7730666666666667,
      "eval_loss": 0.7543359398841858,
      "eval_runtime": 11.4439,
      "eval_samples_per_second": 327.686,
      "eval_steps_per_second": 5.156,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.3225,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7864,
      "eval_loss": 0.7327266931533813,
      "eval_runtime": 11.1871,
      "eval_samples_per_second": 335.207,
      "eval_steps_per_second": 5.274,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.18768703937530518,
      "learning_rate": 0.0003,
      "loss": 0.2683,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.776,
      "eval_loss": 0.7974342703819275,
      "eval_runtime": 11.5279,
      "eval_samples_per_second": 325.296,
      "eval_steps_per_second": 5.118,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.2332,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7890666666666667,
      "eval_loss": 0.7680100202560425,
      "eval_runtime": 11.7548,
      "eval_samples_per_second": 319.019,
      "eval_steps_per_second": 5.019,
      "step": 2664
    }
  ],
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}