trainer_state.json

{
  "best_metric": 0.8909333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0720/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": 2.3952786922454834,
      "learning_rate": 6.125e-05,
      "loss": 1.5349,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8122666666666667,
      "eval_loss": 0.6478140354156494,
      "eval_runtime": 10.3467,
      "eval_samples_per_second": 362.436,
      "eval_steps_per_second": 5.702,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.193082809448242,
      "learning_rate": 5.2499999999999995e-05,
      "loss": 0.4432,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8354666666666667,
      "eval_loss": 0.5375003218650818,
      "eval_runtime": 10.7625,
      "eval_samples_per_second": 348.432,
      "eval_steps_per_second": 5.482,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.25942882895469666,
      "learning_rate": 4.374999999999999e-05,
      "loss": 0.2289,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8712,
      "eval_loss": 0.40710774064064026,
      "eval_runtime": 11.8043,
      "eval_samples_per_second": 317.682,
      "eval_steps_per_second": 4.998,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2744712829589844,
      "learning_rate": 3.5e-05,
      "loss": 0.1159,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.41876643896102905,
      "eval_runtime": 10.7933,
      "eval_samples_per_second": 347.439,
      "eval_steps_per_second": 5.466,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 2.627627627627627e-05,
      "loss": 0.054,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.40753448009490967,
      "eval_runtime": 10.488,
      "eval_samples_per_second": 357.55,
      "eval_steps_per_second": 5.625,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.23905228078365326,
      "learning_rate": 1.7526276276276272e-05,
      "loss": 0.0222,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.4272323548793793,
      "eval_runtime": 10.9526,
      "eval_samples_per_second": 342.385,
      "eval_steps_per_second": 5.387,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.46245670318603516,
      "learning_rate": 8.776276276276276e-06,
      "loss": 0.0078,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.41905146837234497,
      "eval_runtime": 10.3058,
      "eval_samples_per_second": 363.873,
      "eval_steps_per_second": 5.725,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.8751598596572876,
      "learning_rate": 2.6276276276276275e-08,
      "loss": 0.0042,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8909333333333334,
      "eval_loss": 0.41687577962875366,
      "eval_runtime": 11.453,
      "eval_samples_per_second": 327.426,
      "eval_steps_per_second": 5.151,
      "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
}