trainer_state.json

{
  "best_metric": 0.9026666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0231/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": 8.006006006006006e-05,
      "loss": 1.645,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7693333333333333,
      "eval_loss": 0.7928568720817566,
      "eval_runtime": 6.9732,
      "eval_samples_per_second": 537.774,
      "eval_steps_per_second": 8.461,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.6986830830574036,
      "learning_rate": 6.0060060060060066e-05,
      "loss": 0.5451,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8317333333333333,
      "eval_loss": 0.5734447836875916,
      "eval_runtime": 6.9735,
      "eval_samples_per_second": 537.75,
      "eval_steps_per_second": 8.461,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.349012851715088,
      "learning_rate": 4.006006006006006e-05,
      "loss": 0.3147,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4580739140510559,
      "eval_runtime": 7.1139,
      "eval_samples_per_second": 527.139,
      "eval_steps_per_second": 8.294,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1548660397529602,
      "learning_rate": 2.006006006006006e-05,
      "loss": 0.171,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.40863150358200073,
      "eval_runtime": 7.0148,
      "eval_samples_per_second": 534.584,
      "eval_steps_per_second": 8.411,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2633299231529236,
      "learning_rate": 6.006006006006006e-08,
      "loss": 0.0914,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9026666666666666,
      "eval_loss": 0.35280799865722656,
      "eval_runtime": 6.9836,
      "eval_samples_per_second": 536.972,
      "eval_steps_per_second": 8.448,
      "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
}