trainer_state.json

{
  "best_metric": 0.9269333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_dino-vitb16/model_idx_0923/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": 17.047550201416016,
      "learning_rate": 6.127627627627627e-05,
      "loss": 0.6937,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8757333333333334,
      "eval_loss": 0.3825688064098358,
      "eval_runtime": 8.5538,
      "eval_samples_per_second": 438.401,
      "eval_steps_per_second": 6.898,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.10709535330533981,
      "learning_rate": 5.252627627627627e-05,
      "loss": 0.228,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8954666666666666,
      "eval_loss": 0.3508569896221161,
      "eval_runtime": 6.0057,
      "eval_samples_per_second": 624.406,
      "eval_steps_per_second": 9.824,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.0197985172271729,
      "learning_rate": 4.377627627627627e-05,
      "loss": 0.1199,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.36267223954200745,
      "eval_runtime": 5.6451,
      "eval_samples_per_second": 664.288,
      "eval_steps_per_second": 10.451,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.0023195950780063868,
      "learning_rate": 3.502627627627627e-05,
      "loss": 0.0714,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9077333333333333,
      "eval_loss": 0.3463919162750244,
      "eval_runtime": 5.6372,
      "eval_samples_per_second": 665.218,
      "eval_steps_per_second": 10.466,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.24614669382572174,
      "learning_rate": 2.627627627627627e-05,
      "loss": 0.0362,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9176,
      "eval_loss": 0.3222881555557251,
      "eval_runtime": 5.6177,
      "eval_samples_per_second": 667.529,
      "eval_steps_per_second": 10.502,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.047190502285957336,
      "learning_rate": 1.7526276276276272e-05,
      "loss": 0.0138,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9226666666666666,
      "eval_loss": 0.3602088689804077,
      "eval_runtime": 5.9098,
      "eval_samples_per_second": 634.535,
      "eval_steps_per_second": 9.983,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.003061029827222228,
      "learning_rate": 8.776276276276276e-06,
      "loss": 0.0031,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.924,
      "eval_loss": 0.34025508165359497,
      "eval_runtime": 5.7129,
      "eval_samples_per_second": 656.414,
      "eval_steps_per_second": 10.328,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.012218507938086987,
      "learning_rate": 2.6276276276276275e-08,
      "loss": 0.0006,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9269333333333334,
      "eval_loss": 0.3283710479736328,
      "eval_runtime": 5.8732,
      "eval_samples_per_second": 638.498,
      "eval_steps_per_second": 10.046,
      "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
}