trainer_state.json

{
  "best_metric": 0.9237333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0806/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": 0.10735572874546051,
      "learning_rate": 5.604204204204204e-05,
      "loss": 0.7426,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8656,
      "eval_loss": 0.45864352583885193,
      "eval_runtime": 11.9751,
      "eval_samples_per_second": 313.149,
      "eval_steps_per_second": 4.927,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 4.212612612612612e-05,
      "loss": 0.2212,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8928,
      "eval_loss": 0.36810216307640076,
      "eval_runtime": 11.3535,
      "eval_samples_per_second": 330.294,
      "eval_steps_per_second": 5.197,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.6789270639419556,
      "learning_rate": 2.8126126126126122e-05,
      "loss": 0.0968,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.3513356149196625,
      "eval_runtime": 10.9541,
      "eval_samples_per_second": 342.337,
      "eval_steps_per_second": 5.386,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0023910775780677795,
      "learning_rate": 1.4126126126126124e-05,
      "loss": 0.0414,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9101333333333333,
      "eval_loss": 0.3560229241847992,
      "eval_runtime": 11.0497,
      "eval_samples_per_second": 339.377,
      "eval_steps_per_second": 5.34,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.5764195919036865,
      "learning_rate": 1.2612612612612613e-07,
      "loss": 0.0135,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9237333333333333,
      "eval_loss": 0.30356672406196594,
      "eval_runtime": 10.8727,
      "eval_samples_per_second": 344.901,
      "eval_steps_per_second": 5.426,
      "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
}