trainer_state.json

{
  "best_metric": 0.9109333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0240/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": 7.881756756756757e-05,
      "loss": 0.8549,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8498666666666667,
      "eval_loss": 0.47439703345298767,
      "eval_runtime": 15.1975,
      "eval_samples_per_second": 246.751,
      "eval_steps_per_second": 3.882,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.4385628700256348,
      "learning_rate": 6.756756756756757e-05,
      "loss": 0.2652,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.4896063804626465,
      "eval_runtime": 15.297,
      "eval_samples_per_second": 245.147,
      "eval_steps_per_second": 3.857,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.14917206764221191,
      "learning_rate": 5.6317567567567566e-05,
      "loss": 0.1349,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.5200441479682922,
      "eval_runtime": 12.8569,
      "eval_samples_per_second": 291.671,
      "eval_steps_per_second": 4.589,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.0010064840316772,
      "learning_rate": 4.5067567567567564e-05,
      "loss": 0.0758,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.5440783500671387,
      "eval_runtime": 12.099,
      "eval_samples_per_second": 309.944,
      "eval_steps_per_second": 4.876,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3554956912994385,
      "learning_rate": 3.381756756756757e-05,
      "loss": 0.0318,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.892,
      "eval_loss": 0.458025723695755,
      "eval_runtime": 11.7719,
      "eval_samples_per_second": 318.556,
      "eval_steps_per_second": 5.012,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.00028387035126797855,
      "learning_rate": 2.256756756756757e-05,
      "loss": 0.0097,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.4794585108757019,
      "eval_runtime": 12.0833,
      "eval_samples_per_second": 310.346,
      "eval_steps_per_second": 4.883,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.05472806841135025,
      "learning_rate": 1.1317567567567568e-05,
      "loss": 0.0024,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9082666666666667,
      "eval_loss": 0.4398139417171478,
      "eval_runtime": 11.9651,
      "eval_samples_per_second": 313.411,
      "eval_steps_per_second": 4.931,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0011066324077546597,
      "learning_rate": 6.756756756756757e-08,
      "loss": 0.001,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9109333333333334,
      "eval_loss": 0.4328260123729706,
      "eval_runtime": 12.2155,
      "eval_samples_per_second": 306.987,
      "eval_steps_per_second": 4.83,
      "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
}