trainer_state.json

{
  "best_metric": 0.8802666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0177/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": 5.945476531982422,
      "learning_rate": 7.878378378378379e-05,
      "loss": 1.7584,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7704,
      "eval_loss": 0.8097118735313416,
      "eval_runtime": 13.5798,
      "eval_samples_per_second": 276.145,
      "eval_steps_per_second": 4.345,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.3444786071777344,
      "learning_rate": 6.753378378378378e-05,
      "loss": 0.5946,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8146666666666667,
      "eval_loss": 0.635450541973114,
      "eval_runtime": 13.0009,
      "eval_samples_per_second": 288.441,
      "eval_steps_per_second": 4.538,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 5.6317567567567566e-05,
      "loss": 0.3424,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.84,
      "eval_loss": 0.5445099472999573,
      "eval_runtime": 13.0321,
      "eval_samples_per_second": 287.752,
      "eval_steps_per_second": 4.527,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.21000507473945618,
      "learning_rate": 4.5067567567567564e-05,
      "loss": 0.1981,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.49323612451553345,
      "eval_runtime": 13.2488,
      "eval_samples_per_second": 283.044,
      "eval_steps_per_second": 4.453,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 3.385135135135135e-05,
      "loss": 0.1161,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.5033565163612366,
      "eval_runtime": 13.3616,
      "eval_samples_per_second": 280.655,
      "eval_steps_per_second": 4.416,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.8436614871025085,
      "learning_rate": 2.2601351351351353e-05,
      "loss": 0.0587,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.48401665687561035,
      "eval_runtime": 13.1577,
      "eval_samples_per_second": 285.004,
      "eval_steps_per_second": 4.484,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.5812382102012634,
      "learning_rate": 1.135135135135135e-05,
      "loss": 0.0289,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8725333333333334,
      "eval_loss": 0.4840221107006073,
      "eval_runtime": 13.122,
      "eval_samples_per_second": 285.78,
      "eval_steps_per_second": 4.496,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.04750360921025276,
      "learning_rate": 1.0135135135135135e-07,
      "loss": 0.0129,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.47388848662376404,
      "eval_runtime": 13.111,
      "eval_samples_per_second": 286.018,
      "eval_steps_per_second": 4.5,
      "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
}