trainer_state.json

{
  "best_metric": 0.9442666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0936/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 4.378753753753754e-05,
      "loss": 0.6788,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.3338634669780731,
      "eval_runtime": 8.2673,
      "eval_samples_per_second": 453.592,
      "eval_steps_per_second": 7.137,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6040836572647095,
      "learning_rate": 3.7537537537537536e-05,
      "loss": 0.1483,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9098666666666667,
      "eval_loss": 0.2804531455039978,
      "eval_runtime": 5.7335,
      "eval_samples_per_second": 654.049,
      "eval_steps_per_second": 10.29,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.0033770001027733088,
      "learning_rate": 3.128753753753754e-05,
      "loss": 0.0577,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9178666666666667,
      "eval_loss": 0.308549165725708,
      "eval_runtime": 6.2703,
      "eval_samples_per_second": 598.06,
      "eval_steps_per_second": 9.409,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.007872807793319225,
      "learning_rate": 2.5037537537537537e-05,
      "loss": 0.0237,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9189333333333334,
      "eval_loss": 0.312144935131073,
      "eval_runtime": 7.2023,
      "eval_samples_per_second": 520.667,
      "eval_steps_per_second": 8.192,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 1.8806306306306307e-05,
      "loss": 0.0131,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9250666666666667,
      "eval_loss": 0.2745247483253479,
      "eval_runtime": 5.6973,
      "eval_samples_per_second": 658.207,
      "eval_steps_per_second": 10.356,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.010893922299146652,
      "learning_rate": 1.2556306306306306e-05,
      "loss": 0.0028,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9384,
      "eval_loss": 0.24089238047599792,
      "eval_runtime": 5.7034,
      "eval_samples_per_second": 657.497,
      "eval_steps_per_second": 10.345,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.0016305410536006093,
      "learning_rate": 6.306306306306306e-06,
      "loss": 0.0006,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9432,
      "eval_loss": 0.23244334757328033,
      "eval_runtime": 5.7905,
      "eval_samples_per_second": 647.616,
      "eval_steps_per_second": 10.189,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.15240362286567688,
      "learning_rate": 5.630630630630631e-08,
      "loss": 0.0003,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9442666666666667,
      "eval_loss": 0.2300555408000946,
      "eval_runtime": 5.7334,
      "eval_samples_per_second": 654.061,
      "eval_steps_per_second": 10.291,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}