trainer_state.json

{
  "best_metric": 0.8082666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0532/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "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": 1.9992,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6477333333333334,
      "eval_loss": 1.2898966073989868,
      "eval_runtime": 7.6396,
      "eval_samples_per_second": 490.865,
      "eval_steps_per_second": 7.723,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.7526,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7549333333333333,
      "eval_loss": 0.8646398186683655,
      "eval_runtime": 6.9127,
      "eval_samples_per_second": 542.477,
      "eval_steps_per_second": 8.535,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.4776,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7877333333333333,
      "eval_loss": 0.7354907393455505,
      "eval_runtime": 6.8735,
      "eval_samples_per_second": 545.574,
      "eval_steps_per_second": 8.584,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 6.350447177886963,
      "learning_rate": 0.0003,
      "loss": 0.3387,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7930666666666667,
      "eval_loss": 0.7302750945091248,
      "eval_runtime": 7.1147,
      "eval_samples_per_second": 527.076,
      "eval_steps_per_second": 8.293,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.8281162977218628,
      "learning_rate": 0.0003,
      "loss": 0.261,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8082666666666667,
      "eval_loss": 0.728571891784668,
      "eval_runtime": 6.9716,
      "eval_samples_per_second": 537.893,
      "eval_steps_per_second": 8.463,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}