trainer_state.json

{
  "best_metric": 0.8653333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0916/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": 0.0002886495269365962,
      "loss": 2.4141,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6386666666666667,
      "eval_loss": 1.1988667249679565,
      "eval_runtime": 6.5614,
      "eval_samples_per_second": 571.525,
      "eval_steps_per_second": 8.992,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.28454849123954773,
      "learning_rate": 0.00025619102456015866,
      "loss": 0.9219,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7504,
      "eval_loss": 0.8495290279388428,
      "eval_runtime": 6.5268,
      "eval_samples_per_second": 574.552,
      "eval_steps_per_second": 9.04,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0002077292077256867,
      "loss": 0.5251,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7946666666666666,
      "eval_loss": 0.6748585104942322,
      "eval_runtime": 6.3591,
      "eval_samples_per_second": 589.706,
      "eval_steps_per_second": 9.278,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 5.876136302947998,
      "learning_rate": 0.00015035378262848306,
      "loss": 0.2801,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8226666666666667,
      "eval_loss": 0.5865784883499146,
      "eval_runtime": 6.7048,
      "eval_samples_per_second": 559.304,
      "eval_steps_per_second": 8.8,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.020033687353134155,
      "learning_rate": 9.29244973331404e-05,
      "loss": 0.1192,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.5580995082855225,
      "eval_runtime": 6.0179,
      "eval_samples_per_second": 623.138,
      "eval_steps_per_second": 9.804,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.2547885775566101,
      "learning_rate": 4.4184439928107014e-05,
      "loss": 0.044,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.5402912497520447,
      "eval_runtime": 6.7424,
      "eval_samples_per_second": 556.182,
      "eval_steps_per_second": 8.751,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.01063931081444025,
      "learning_rate": 1.1553842323578666e-05,
      "loss": 0.0132,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.5106397867202759,
      "eval_runtime": 7.1554,
      "eval_samples_per_second": 524.079,
      "eval_steps_per_second": 8.246,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.232347249984741,
      "learning_rate": 4.172077409214658e-10,
      "loss": 0.0064,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8653333333333333,
      "eval_loss": 0.5077541470527649,
      "eval_runtime": 6.0039,
      "eval_samples_per_second": 624.591,
      "eval_steps_per_second": 9.827,
      "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
}