trainer_state.json

{
  "best_metric": 0.8904,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0535/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": 4.003003003003003e-05,
      "loss": 2.1784,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.772,
      "eval_loss": 1.033983588218689,
      "eval_runtime": 7.9583,
      "eval_samples_per_second": 471.205,
      "eval_steps_per_second": 7.414,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.4331207275390625,
      "learning_rate": 3.0030030030030033e-05,
      "loss": 0.7027,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8381333333333333,
      "eval_loss": 0.6028279662132263,
      "eval_runtime": 6.7795,
      "eval_samples_per_second": 553.139,
      "eval_steps_per_second": 8.703,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 2.006006006006006e-05,
      "loss": 0.4008,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8584,
      "eval_loss": 0.4869285523891449,
      "eval_runtime": 6.4692,
      "eval_samples_per_second": 579.67,
      "eval_steps_per_second": 9.12,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.0554139614105225,
      "learning_rate": 1.006006006006006e-05,
      "loss": 0.2466,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.4085443913936615,
      "eval_runtime": 6.3375,
      "eval_samples_per_second": 591.719,
      "eval_steps_per_second": 9.31,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.5750203132629395,
      "learning_rate": 6.006006006006006e-08,
      "loss": 0.1618,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.39886024594306946,
      "eval_runtime": 6.4203,
      "eval_samples_per_second": 584.087,
      "eval_steps_per_second": 9.19,
      "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
}