trainer_state.json

{
  "best_metric": 0.6306666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0542/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": 0.00028523292208850295,
      "loss": 3.1469,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.244,
      "eval_loss": 2.820418119430542,
      "eval_runtime": 12.4722,
      "eval_samples_per_second": 300.668,
      "eval_steps_per_second": 4.731,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.000243839242887935,
      "loss": 2.4843,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.37066666666666664,
      "eval_loss": 2.277475357055664,
      "eval_runtime": 12.3351,
      "eval_samples_per_second": 304.011,
      "eval_steps_per_second": 4.783,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.537004470825195,
      "learning_rate": 0.00018377220450662088,
      "loss": 1.958,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4517333333333333,
      "eval_loss": 1.9610399007797241,
      "eval_runtime": 12.2398,
      "eval_samples_per_second": 306.378,
      "eval_steps_per_second": 4.82,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 7.14243221282959,
      "learning_rate": 0.00011701616683386335,
      "loss": 1.4682,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5448,
      "eval_loss": 1.6052461862564087,
      "eval_runtime": 12.347,
      "eval_samples_per_second": 303.717,
      "eval_steps_per_second": 4.778,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": Infinity,
      "learning_rate": 5.695146199689343e-05,
      "loss": 0.985,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6053333333333333,
      "eval_loss": 1.4043488502502441,
      "eval_runtime": 12.2226,
      "eval_samples_per_second": 306.808,
      "eval_steps_per_second": 4.827,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": Infinity,
      "learning_rate": 1.5207490629826985e-05,
      "loss": 0.5942,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6306666666666667,
      "eval_loss": 1.325668215751648,
      "eval_runtime": 12.146,
      "eval_samples_per_second": 308.744,
      "eval_steps_per_second": 4.858,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}