trainer_state.json

{
  "best_metric": 0.9173333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0207/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": 8.097506523132324,
      "learning_rate": 4.525311529382417e-05,
      "loss": 0.736,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8368,
      "eval_loss": 0.5320413112640381,
      "eval_runtime": 7.5879,
      "eval_samples_per_second": 494.208,
      "eval_steps_per_second": 7.776,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 81.07645416259766,
      "learning_rate": 3.277027342564428e-05,
      "loss": 0.183,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.39614158868789673,
      "eval_runtime": 6.3582,
      "eval_samples_per_second": 589.786,
      "eval_steps_per_second": 9.279,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0294552780687809,
      "learning_rate": 1.731945121074836e-05,
      "loss": 0.0517,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.37566396594047546,
      "eval_runtime": 11.7892,
      "eval_samples_per_second": 318.088,
      "eval_steps_per_second": 5.005,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0723167136311531,
      "learning_rate": 4.802337581094707e-06,
      "loss": 0.0082,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9152,
      "eval_loss": 0.33903586864471436,
      "eval_runtime": 6.429,
      "eval_samples_per_second": 583.297,
      "eval_steps_per_second": 9.177,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.002934150630608201,
      "learning_rate": 4.4502166460236305e-11,
      "loss": 0.002,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9173333333333333,
      "eval_loss": 0.3350743353366852,
      "eval_runtime": 6.1943,
      "eval_samples_per_second": 605.392,
      "eval_steps_per_second": 9.525,
      "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
}