trainer_state.json

{
  "best_metric": 0.5330666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0285/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": 0.0005,
      "loss": 3.1407,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2504,
      "eval_loss": 2.7578933238983154,
      "eval_runtime": 6.0778,
      "eval_samples_per_second": 617.002,
      "eval_steps_per_second": 9.707,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 2.4375,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.35706666666666664,
      "eval_loss": 2.3590962886810303,
      "eval_runtime": 5.6743,
      "eval_samples_per_second": 660.873,
      "eval_steps_per_second": 10.398,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0005,
      "loss": 1.9831,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4301333333333333,
      "eval_loss": 2.051231861114502,
      "eval_runtime": 5.7525,
      "eval_samples_per_second": 651.886,
      "eval_steps_per_second": 10.256,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 16.156476974487305,
      "learning_rate": 0.0005,
      "loss": 1.6394,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.47146666666666665,
      "eval_loss": 1.9019935131072998,
      "eval_runtime": 5.3776,
      "eval_samples_per_second": 697.335,
      "eval_steps_per_second": 10.971,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.77993106842041,
      "learning_rate": 0.0005,
      "loss": 1.3526,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5330666666666667,
      "eval_loss": 1.6187279224395752,
      "eval_runtime": 5.4033,
      "eval_samples_per_second": 694.026,
      "eval_steps_per_second": 10.919,
      "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
}