trainer_state.json

{
  "best_metric": 0.8605333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0332/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": 7e-05,
      "loss": 1.5656,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7794666666666666,
      "eval_loss": 0.7875555753707886,
      "eval_runtime": 5.8822,
      "eval_samples_per_second": 637.518,
      "eval_steps_per_second": 10.03,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.40973496437072754,
      "learning_rate": 7e-05,
      "loss": 0.4576,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8509333333333333,
      "eval_loss": 0.5194773077964783,
      "eval_runtime": 5.6319,
      "eval_samples_per_second": 665.847,
      "eval_steps_per_second": 10.476,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.2497,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.47009211778640747,
      "eval_runtime": 5.6959,
      "eval_samples_per_second": 658.371,
      "eval_steps_per_second": 10.358,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.25619375705718994,
      "learning_rate": 7e-05,
      "loss": 0.1554,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.49834567308425903,
      "eval_runtime": 5.706,
      "eval_samples_per_second": 657.207,
      "eval_steps_per_second": 10.34,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.25673654675483704,
      "learning_rate": 7e-05,
      "loss": 0.1065,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.5289484262466431,
      "eval_runtime": 5.7919,
      "eval_samples_per_second": 647.451,
      "eval_steps_per_second": 10.187,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.3245350420475006,
      "learning_rate": 7e-05,
      "loss": 0.0788,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.5675150752067566,
      "eval_runtime": 5.7027,
      "eval_samples_per_second": 657.578,
      "eval_steps_per_second": 10.346,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.1113765686750412,
      "learning_rate": 7e-05,
      "loss": 0.0632,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.576711118221283,
      "eval_runtime": 5.7037,
      "eval_samples_per_second": 657.468,
      "eval_steps_per_second": 10.344,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.2091275304555893,
      "learning_rate": 7e-05,
      "loss": 0.0455,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.5950403213500977,
      "eval_runtime": 6.0222,
      "eval_samples_per_second": 622.693,
      "eval_steps_per_second": 9.797,
      "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
}