trainer_state.json

{
  "best_metric": 0.5552,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0102/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": 0.0003,
      "loss": 3.2689,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2557333333333333,
      "eval_loss": 2.8198866844177246,
      "eval_runtime": 6.6088,
      "eval_samples_per_second": 567.423,
      "eval_steps_per_second": 8.927,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 2.4583,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.37866666666666665,
      "eval_loss": 2.2196974754333496,
      "eval_runtime": 6.4783,
      "eval_samples_per_second": 578.854,
      "eval_steps_per_second": 9.107,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0003,
      "loss": 1.9148,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4845333333333333,
      "eval_loss": 1.8011943101882935,
      "eval_runtime": 6.3973,
      "eval_samples_per_second": 586.185,
      "eval_steps_per_second": 9.223,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 10.146971702575684,
      "learning_rate": 0.0003,
      "loss": 1.4746,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5061333333333333,
      "eval_loss": 1.7187739610671997,
      "eval_runtime": 6.5316,
      "eval_samples_per_second": 574.133,
      "eval_steps_per_second": 9.033,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 21.75577163696289,
      "learning_rate": 0.0003,
      "loss": 1.1232,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5552,
      "eval_loss": 1.5839169025421143,
      "eval_runtime": 6.5718,
      "eval_samples_per_second": 570.624,
      "eval_steps_per_second": 8.978,
      "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
}