trainer_state.json

{
  "best_metric": 0.5285333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0193/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": 0.0005,
      "loss": 3.117,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.25733333333333336,
      "eval_loss": 2.7036898136138916,
      "eval_runtime": 11.5207,
      "eval_samples_per_second": 325.502,
      "eval_steps_per_second": 5.121,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.4474,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3624,
      "eval_loss": 2.2774360179901123,
      "eval_runtime": 11.8847,
      "eval_samples_per_second": 315.532,
      "eval_steps_per_second": 4.964,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 10.66565227508545,
      "learning_rate": 0.0005,
      "loss": 2.0859,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.40826666666666667,
      "eval_loss": 2.0522232055664062,
      "eval_runtime": 11.184,
      "eval_samples_per_second": 335.301,
      "eval_steps_per_second": 5.275,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 12.604762077331543,
      "learning_rate": 0.0005,
      "loss": 1.8158,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4712,
      "eval_loss": 1.8667511940002441,
      "eval_runtime": 11.1365,
      "eval_samples_per_second": 336.731,
      "eval_steps_per_second": 5.298,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 7.917346477508545,
      "learning_rate": 0.0005,
      "loss": 1.5848,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.48773333333333335,
      "eval_loss": 1.8190765380859375,
      "eval_runtime": 11.0091,
      "eval_samples_per_second": 340.627,
      "eval_steps_per_second": 5.359,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.334887504577637,
      "learning_rate": 0.0005,
      "loss": 1.3668,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5285333333333333,
      "eval_loss": 1.6919288635253906,
      "eval_runtime": 11.4047,
      "eval_samples_per_second": 328.813,
      "eval_steps_per_second": 5.173,
      "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
}