trainer_state.json

{
  "best_metric": 0.3101333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0694/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": 3.4505,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.18106666666666665,
      "eval_loss": 3.135206460952759,
      "eval_runtime": 12.4144,
      "eval_samples_per_second": 302.068,
      "eval_steps_per_second": 4.753,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 18.843936920166016,
      "learning_rate": 0.0005,
      "loss": 2.9839,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.21733333333333332,
      "eval_loss": 2.9639902114868164,
      "eval_runtime": 12.3523,
      "eval_samples_per_second": 303.588,
      "eval_steps_per_second": 4.776,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 14.786829948425293,
      "learning_rate": 0.0005,
      "loss": 2.8032,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.2538666666666667,
      "eval_loss": 2.782071828842163,
      "eval_runtime": 12.3474,
      "eval_samples_per_second": 303.707,
      "eval_steps_per_second": 4.778,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 21.32419776916504,
      "learning_rate": 0.0005,
      "loss": 2.6916,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.244,
      "eval_loss": 2.789285898208618,
      "eval_runtime": 11.603,
      "eval_samples_per_second": 323.192,
      "eval_steps_per_second": 5.085,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 19.519277572631836,
      "learning_rate": 0.0005,
      "loss": 2.5948,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.27946666666666664,
      "eval_loss": 2.661184072494507,
      "eval_runtime": 12.2369,
      "eval_samples_per_second": 306.449,
      "eval_steps_per_second": 4.821,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 26.100261688232422,
      "learning_rate": 0.0005,
      "loss": 2.5044,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.3101333333333333,
      "eval_loss": 2.5478625297546387,
      "eval_runtime": 12.3498,
      "eval_samples_per_second": 303.649,
      "eval_steps_per_second": 4.777,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}