trainer_state.json

{
  "best_metric": 0.9322666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0157/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": 5e-05,
      "loss": 0.8408,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.92,
      "eval_loss": 0.26172834634780884,
      "eval_runtime": 6.1716,
      "eval_samples_per_second": 607.621,
      "eval_steps_per_second": 9.56,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.1382649689912796,
      "learning_rate": 5e-05,
      "loss": 0.1411,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.24265696108341217,
      "eval_runtime": 6.617,
      "eval_samples_per_second": 566.724,
      "eval_steps_per_second": 8.916,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.7051064372062683,
      "learning_rate": 5e-05,
      "loss": 0.0697,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9245333333333333,
      "eval_loss": 0.27153849601745605,
      "eval_runtime": 6.1056,
      "eval_samples_per_second": 614.189,
      "eval_steps_per_second": 9.663,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.05140962824225426,
      "learning_rate": 5e-05,
      "loss": 0.0432,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9277333333333333,
      "eval_loss": 0.2559788227081299,
      "eval_runtime": 6.3363,
      "eval_samples_per_second": 591.833,
      "eval_steps_per_second": 9.311,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.21602562069892883,
      "learning_rate": 5e-05,
      "loss": 0.0283,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9304,
      "eval_loss": 0.27624574303627014,
      "eval_runtime": 5.9704,
      "eval_samples_per_second": 628.094,
      "eval_steps_per_second": 9.882,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.5941274166107178,
      "learning_rate": 5e-05,
      "loss": 0.0188,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9322666666666667,
      "eval_loss": 0.30486539006233215,
      "eval_runtime": 6.1031,
      "eval_samples_per_second": 614.447,
      "eval_steps_per_second": 9.667,
      "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
}