trainer_state.json

{
  "best_metric": 0.7512,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0586/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": 2.8932,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.39466666666666667,
      "eval_loss": 2.1456875801086426,
      "eval_runtime": 5.7113,
      "eval_samples_per_second": 656.593,
      "eval_steps_per_second": 10.33,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 10.598609924316406,
      "learning_rate": 0.0003,
      "loss": 1.3933,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6192,
      "eval_loss": 1.2717610597610474,
      "eval_runtime": 5.7561,
      "eval_samples_per_second": 651.478,
      "eval_steps_per_second": 10.25,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 7.238413333892822,
      "learning_rate": 0.0003,
      "loss": 0.8159,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7082666666666667,
      "eval_loss": 0.9565859436988831,
      "eval_runtime": 6.1272,
      "eval_samples_per_second": 612.024,
      "eval_steps_per_second": 9.629,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 0.5478,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7501333333333333,
      "eval_loss": 0.8165543675422668,
      "eval_runtime": 5.917,
      "eval_samples_per_second": 633.767,
      "eval_steps_per_second": 9.971,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.21206149458885193,
      "learning_rate": 0.0003,
      "loss": 0.3712,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7512,
      "eval_loss": 0.8557840585708618,
      "eval_runtime": 5.9251,
      "eval_samples_per_second": 632.901,
      "eval_steps_per_second": 9.958,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}