trainer_state.json

{
  "best_metric": 0.2816,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0104/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": 0.0005,
      "loss": 3.5563,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1736,
      "eval_loss": 3.1686997413635254,
      "eval_runtime": 11.9405,
      "eval_samples_per_second": 314.058,
      "eval_steps_per_second": 4.941,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 3.1185,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.21066666666666667,
      "eval_loss": 3.0196797847747803,
      "eval_runtime": 11.9038,
      "eval_samples_per_second": 315.024,
      "eval_steps_per_second": 4.956,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 18.367876052856445,
      "learning_rate": 0.0005,
      "loss": 2.9511,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.22693333333333332,
      "eval_loss": 2.9659605026245117,
      "eval_runtime": 11.9456,
      "eval_samples_per_second": 313.924,
      "eval_steps_per_second": 4.939,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 14.879671096801758,
      "learning_rate": 0.0005,
      "loss": 2.8171,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.24853333333333333,
      "eval_loss": 2.826829195022583,
      "eval_runtime": 11.9493,
      "eval_samples_per_second": 313.827,
      "eval_steps_per_second": 4.938,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 17.593599319458008,
      "learning_rate": 0.0005,
      "loss": 2.6921,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.26426666666666665,
      "eval_loss": 2.719712257385254,
      "eval_runtime": 11.9184,
      "eval_samples_per_second": 314.64,
      "eval_steps_per_second": 4.95,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 9.95388126373291,
      "learning_rate": 0.0005,
      "loss": 2.5714,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.2816,
      "eval_loss": 2.6526269912719727,
      "eval_runtime": 12.0751,
      "eval_samples_per_second": 310.556,
      "eval_steps_per_second": 4.886,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}