trainer_state.json

{
  "best_metric": 0.8970666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0126/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": 2.886495269365963e-05,
      "loss": 2.2839,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7421333333333333,
      "eval_loss": 1.1323415040969849,
      "eval_runtime": 6.2482,
      "eval_samples_per_second": 600.174,
      "eval_steps_per_second": 9.443,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 17.75069236755371,
      "learning_rate": 2.5619102456015873e-05,
      "loss": 0.7793,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8290666666666666,
      "eval_loss": 0.6547415852546692,
      "eval_runtime": 6.3082,
      "eval_samples_per_second": 594.46,
      "eval_steps_per_second": 9.353,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 20.086994171142578,
      "learning_rate": 2.0756590131847176e-05,
      "loss": 0.43,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.4920029938220978,
      "eval_runtime": 6.5611,
      "eval_samples_per_second": 571.548,
      "eval_steps_per_second": 8.992,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6329180598258972,
      "learning_rate": 1.5017689143724237e-05,
      "loss": 0.2672,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.872,
      "eval_loss": 0.45422878861427307,
      "eval_runtime": 6.5126,
      "eval_samples_per_second": 575.803,
      "eval_steps_per_second": 9.059,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.9419646263122559,
      "learning_rate": 9.276095143821772e-06,
      "loss": 0.1663,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.4136982858181,
      "eval_runtime": 6.7494,
      "eval_samples_per_second": 555.606,
      "eval_steps_per_second": 8.742,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.0547443628311157,
      "learning_rate": 4.4059137709457125e-06,
      "loss": 0.1069,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8917333333333334,
      "eval_loss": 0.3866594731807709,
      "eval_runtime": 6.386,
      "eval_samples_per_second": 587.218,
      "eval_steps_per_second": 9.239,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 1.3695801496505737,
      "learning_rate": 1.1485859908123557e-06,
      "loss": 0.0734,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8925333333333333,
      "eval_loss": 0.3858962953090668,
      "eval_runtime": 6.2191,
      "eval_samples_per_second": 602.981,
      "eval_steps_per_second": 9.487,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.49813300371170044,
      "learning_rate": 1.04301971493026e-11,
      "loss": 0.0606,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8970666666666667,
      "eval_loss": 0.379980206489563,
      "eval_runtime": 6.5483,
      "eval_samples_per_second": 572.668,
      "eval_steps_per_second": 9.01,
      "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
}