trainer_state.json

{
  "best_metric": 0.8973333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0925/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": 8.006006006006006e-05,
      "loss": 0.8662,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8386666666666667,
      "eval_loss": 0.5226887464523315,
      "eval_runtime": 10.7803,
      "eval_samples_per_second": 347.856,
      "eval_steps_per_second": 5.473,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7.00900900900901e-05,
      "loss": 0.2923,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8445333333333334,
      "eval_loss": 0.5077577233314514,
      "eval_runtime": 10.8568,
      "eval_samples_per_second": 345.405,
      "eval_steps_per_second": 5.434,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.10728190094232559,
      "learning_rate": 6.00900900900901e-05,
      "loss": 0.1441,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8528,
      "eval_loss": 0.5157790184020996,
      "eval_runtime": 11.2745,
      "eval_samples_per_second": 332.61,
      "eval_steps_per_second": 5.233,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1070307120680809,
      "learning_rate": 5.0090090090090095e-05,
      "loss": 0.0816,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.876,
      "eval_loss": 0.4824523329734802,
      "eval_runtime": 10.7984,
      "eval_samples_per_second": 347.274,
      "eval_steps_per_second": 5.464,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.039365917444229126,
      "learning_rate": 4.0090090090090096e-05,
      "loss": 0.0411,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.5307037234306335,
      "eval_runtime": 10.4701,
      "eval_samples_per_second": 358.164,
      "eval_steps_per_second": 5.635,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.0007872133865021169,
      "learning_rate": 3.0090090090090093e-05,
      "loss": 0.0207,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8768,
      "eval_loss": 0.548275887966156,
      "eval_runtime": 10.8599,
      "eval_samples_per_second": 345.306,
      "eval_steps_per_second": 5.433,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.2063124179840088,
      "learning_rate": 2.009009009009009e-05,
      "loss": 0.0055,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.49143943190574646,
      "eval_runtime": 10.8471,
      "eval_samples_per_second": 345.715,
      "eval_steps_per_second": 5.439,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.010951201431453228,
      "learning_rate": 1.009009009009009e-05,
      "loss": 0.0022,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.47764137387275696,
      "eval_runtime": 10.7084,
      "eval_samples_per_second": 350.193,
      "eval_steps_per_second": 5.51,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}