trainer_state.json

{
  "best_metric": 0.856,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0555/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": 9e-05,
      "loss": 2.6165,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6912,
      "eval_loss": 1.127916693687439,
      "eval_runtime": 8.5652,
      "eval_samples_per_second": 437.82,
      "eval_steps_per_second": 6.888,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.7542,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8093333333333333,
      "eval_loss": 0.632706344127655,
      "eval_runtime": 7.6555,
      "eval_samples_per_second": 489.846,
      "eval_steps_per_second": 7.707,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.8318922519683838,
      "learning_rate": 9e-05,
      "loss": 0.4292,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.828,
      "eval_loss": 0.5654412508010864,
      "eval_runtime": 7.2915,
      "eval_samples_per_second": 514.298,
      "eval_steps_per_second": 8.092,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.4031338691711426,
      "learning_rate": 9e-05,
      "loss": 0.2992,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8352,
      "eval_loss": 0.5648061037063599,
      "eval_runtime": 7.3401,
      "eval_samples_per_second": 510.894,
      "eval_steps_per_second": 8.038,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6059828996658325,
      "learning_rate": 9e-05,
      "loss": 0.2168,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.5275124907493591,
      "eval_runtime": 7.5619,
      "eval_samples_per_second": 495.908,
      "eval_steps_per_second": 7.802,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.26083213090896606,
      "learning_rate": 9e-05,
      "loss": 0.1637,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.49012404680252075,
      "eval_runtime": 8.2075,
      "eval_samples_per_second": 456.897,
      "eval_steps_per_second": 7.189,
      "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
}