trainer_state.json

{
  "best_metric": 0.5448,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0276/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "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": 0.0003,
      "loss": 3.1985,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.25893333333333335,
      "eval_loss": 2.693833827972412,
      "eval_runtime": 7.8761,
      "eval_samples_per_second": 476.123,
      "eval_steps_per_second": 7.491,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.4958,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3552,
      "eval_loss": 2.308619499206543,
      "eval_runtime": 7.4624,
      "eval_samples_per_second": 502.522,
      "eval_steps_per_second": 7.906,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.0717,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.45466666666666666,
      "eval_loss": 1.9239963293075562,
      "eval_runtime": 6.9781,
      "eval_samples_per_second": 537.395,
      "eval_steps_per_second": 8.455,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 22.490310668945312,
      "learning_rate": 0.0003,
      "loss": 1.7218,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4826666666666667,
      "eval_loss": 1.8519562482833862,
      "eval_runtime": 6.9649,
      "eval_samples_per_second": 538.417,
      "eval_steps_per_second": 8.471,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 14.53004264831543,
      "learning_rate": 0.0003,
      "loss": 1.4357,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5205333333333333,
      "eval_loss": 1.6878550052642822,
      "eval_runtime": 6.7954,
      "eval_samples_per_second": 551.84,
      "eval_steps_per_second": 8.682,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 18.503171920776367,
      "learning_rate": 0.0003,
      "loss": 1.1862,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5448,
      "eval_loss": 1.602820873260498,
      "eval_runtime": 7.5982,
      "eval_samples_per_second": 493.537,
      "eval_steps_per_second": 7.765,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 1998,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}