trainer_state.json

{
  "best_metric": 0.7877333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0522/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": 1.6291700601577759,
      "learning_rate": 0.0003,
      "loss": 1.8965,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7013333333333334,
      "eval_loss": 1.0003502368927002,
      "eval_runtime": 12.0152,
      "eval_samples_per_second": 312.106,
      "eval_steps_per_second": 4.91,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.6998140811920166,
      "learning_rate": 0.0003,
      "loss": 0.7485,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7325333333333334,
      "eval_loss": 0.8788936138153076,
      "eval_runtime": 11.9055,
      "eval_samples_per_second": 314.98,
      "eval_steps_per_second": 4.956,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.4721040725708,
      "learning_rate": 0.0003,
      "loss": 0.4451,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7762666666666667,
      "eval_loss": 0.7424725890159607,
      "eval_runtime": 11.8979,
      "eval_samples_per_second": 315.181,
      "eval_steps_per_second": 4.959,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.2957,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7744,
      "eval_loss": 0.7911914587020874,
      "eval_runtime": 11.762,
      "eval_samples_per_second": 318.824,
      "eval_steps_per_second": 5.016,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.7387237548828125,
      "learning_rate": 0.0003,
      "loss": 0.2153,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7858666666666667,
      "eval_loss": 0.7774655222892761,
      "eval_runtime": 11.8589,
      "eval_samples_per_second": 316.217,
      "eval_steps_per_second": 4.975,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 6.813511848449707,
      "learning_rate": 0.0003,
      "loss": 0.1805,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7877333333333333,
      "eval_loss": 0.8293803334236145,
      "eval_runtime": 11.8774,
      "eval_samples_per_second": 315.727,
      "eval_steps_per_second": 4.967,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}