trainer_state.json

{
  "best_metric": 0.848,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0368/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": 5e-05,
      "loss": 2.0008,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7672,
      "eval_loss": 0.9127269983291626,
      "eval_runtime": 6.018,
      "eval_samples_per_second": 623.134,
      "eval_steps_per_second": 9.804,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 5e-05,
      "loss": 0.665,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8152,
      "eval_loss": 0.6254153251647949,
      "eval_runtime": 5.9912,
      "eval_samples_per_second": 625.917,
      "eval_steps_per_second": 9.848,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.051182508468628,
      "learning_rate": 5e-05,
      "loss": 0.4086,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8301333333333333,
      "eval_loss": 0.5696828365325928,
      "eval_runtime": 6.1871,
      "eval_samples_per_second": 606.103,
      "eval_steps_per_second": 9.536,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.2470016479492188,
      "learning_rate": 5e-05,
      "loss": 0.282,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8301333333333333,
      "eval_loss": 0.5664248466491699,
      "eval_runtime": 6.1746,
      "eval_samples_per_second": 607.328,
      "eval_steps_per_second": 9.555,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.16398170590400696,
      "learning_rate": 5e-05,
      "loss": 0.1943,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8456,
      "eval_loss": 0.5133583545684814,
      "eval_runtime": 6.5985,
      "eval_samples_per_second": 568.315,
      "eval_steps_per_second": 8.941,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.4426308870315552,
      "learning_rate": 5e-05,
      "loss": 0.1477,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.848,
      "eval_loss": 0.5239871144294739,
      "eval_runtime": 6.0782,
      "eval_samples_per_second": 616.958,
      "eval_steps_per_second": 9.707,
      "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
}