trainer_state.json

{
  "best_metric": 0.8874666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0452/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 21.445119857788086,
      "learning_rate": 0.0001,
      "loss": 1.3823,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8208,
      "eval_loss": 0.6344112157821655,
      "eval_runtime": 5.6398,
      "eval_samples_per_second": 664.918,
      "eval_steps_per_second": 10.461,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.4146175384521484,
      "learning_rate": 0.0001,
      "loss": 0.3955,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8624,
      "eval_loss": 0.45602989196777344,
      "eval_runtime": 5.5884,
      "eval_samples_per_second": 671.032,
      "eval_steps_per_second": 10.558,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.8574941754341125,
      "learning_rate": 0.0001,
      "loss": 0.2069,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4488532543182373,
      "eval_runtime": 5.5486,
      "eval_samples_per_second": 675.848,
      "eval_steps_per_second": 10.633,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.3846279382705688,
      "learning_rate": 0.0001,
      "loss": 0.1269,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8834666666666666,
      "eval_loss": 0.45313817262649536,
      "eval_runtime": 6.2686,
      "eval_samples_per_second": 598.221,
      "eval_steps_per_second": 9.412,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.04474795237183571,
      "learning_rate": 0.0001,
      "loss": 0.0913,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8866666666666667,
      "eval_loss": 0.421602725982666,
      "eval_runtime": 5.9282,
      "eval_samples_per_second": 632.567,
      "eval_steps_per_second": 9.952,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 7.448238372802734,
      "learning_rate": 0.0001,
      "loss": 0.0729,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.4562315344810486,
      "eval_runtime": 5.6071,
      "eval_samples_per_second": 668.792,
      "eval_steps_per_second": 10.522,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.38349202275276184,
      "learning_rate": 0.0001,
      "loss": 0.0642,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.4792778789997101,
      "eval_runtime": 5.583,
      "eval_samples_per_second": 671.679,
      "eval_steps_per_second": 10.568,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}