trainer_state.json

{
  "best_metric": 0.3933333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0926/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 3.2033,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2432,
      "eval_loss": 2.816586971282959,
      "eval_runtime": 10.7105,
      "eval_samples_per_second": 350.123,
      "eval_steps_per_second": 5.509,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.6682,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.2677333333333333,
      "eval_loss": 2.6912827491760254,
      "eval_runtime": 10.735,
      "eval_samples_per_second": 349.325,
      "eval_steps_per_second": 5.496,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 27.529033660888672,
      "learning_rate": 0.0003,
      "loss": 2.4309,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.3152,
      "eval_loss": 2.485410213470459,
      "eval_runtime": 10.4691,
      "eval_samples_per_second": 358.198,
      "eval_steps_per_second": 5.636,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 35.79300308227539,
      "learning_rate": 0.0003,
      "loss": 2.2335,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.3552,
      "eval_loss": 2.288602113723755,
      "eval_runtime": 10.4933,
      "eval_samples_per_second": 357.372,
      "eval_steps_per_second": 5.623,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 23.59615135192871,
      "learning_rate": 0.0003,
      "loss": 2.075,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.3933333333333333,
      "eval_loss": 2.1898252964019775,
      "eval_runtime": 10.4573,
      "eval_samples_per_second": 358.602,
      "eval_steps_per_second": 5.642,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}