trainer_state.json

{
  "best_metric": 0.8405333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0944/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": 0.0001,
      "loss": 0.8839,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8165333333333333,
      "eval_loss": 0.6082785725593567,
      "eval_runtime": 12.6222,
      "eval_samples_per_second": 297.095,
      "eval_steps_per_second": 4.674,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.3675,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8365333333333334,
      "eval_loss": 0.5452620387077332,
      "eval_runtime": 12.4463,
      "eval_samples_per_second": 301.296,
      "eval_steps_per_second": 4.74,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.3448461294174194,
      "learning_rate": 0.0001,
      "loss": 0.2389,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8357333333333333,
      "eval_loss": 0.6131463646888733,
      "eval_runtime": 12.6087,
      "eval_samples_per_second": 297.414,
      "eval_steps_per_second": 4.679,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0001,
      "loss": 0.1822,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8232,
      "eval_loss": 0.6882852911949158,
      "eval_runtime": 12.6392,
      "eval_samples_per_second": 296.697,
      "eval_steps_per_second": 4.668,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2690642774105072,
      "learning_rate": 0.0001,
      "loss": 0.1517,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8405333333333334,
      "eval_loss": 0.6348048448562622,
      "eval_runtime": 12.6794,
      "eval_samples_per_second": 295.755,
      "eval_steps_per_second": 4.653,
      "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
}