trainer_state.json

{
  "best_metric": 0.31866666666666665,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0251/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": 19.144601821899414,
      "learning_rate": 0.0005,
      "loss": 3.4476,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.192,
      "eval_loss": 3.0185365676879883,
      "eval_runtime": 6.1965,
      "eval_samples_per_second": 605.183,
      "eval_steps_per_second": 9.522,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 21.085098266601562,
      "learning_rate": 0.0005,
      "loss": 2.961,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.23493333333333333,
      "eval_loss": 2.8776988983154297,
      "eval_runtime": 5.9337,
      "eval_samples_per_second": 631.981,
      "eval_steps_per_second": 9.943,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 15.725351333618164,
      "learning_rate": 0.0005,
      "loss": 2.8214,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.25333333333333335,
      "eval_loss": 2.7795941829681396,
      "eval_runtime": 5.8665,
      "eval_samples_per_second": 639.223,
      "eval_steps_per_second": 10.057,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 19.15825653076172,
      "learning_rate": 0.0005,
      "loss": 2.6989,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.30533333333333335,
      "eval_loss": 2.6084768772125244,
      "eval_runtime": 6.1289,
      "eval_samples_per_second": 611.856,
      "eval_steps_per_second": 9.627,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 15.900651931762695,
      "learning_rate": 0.0005,
      "loss": 2.5802,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.31866666666666665,
      "eval_loss": 2.542264699935913,
      "eval_runtime": 5.8743,
      "eval_samples_per_second": 638.369,
      "eval_steps_per_second": 10.044,
      "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
}