trainer_state.json

{
  "best_metric": 0.8730666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0184/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "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": 1.6215,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8053333333333333,
      "eval_loss": 0.7093351483345032,
      "eval_runtime": 6.8747,
      "eval_samples_per_second": 545.481,
      "eval_steps_per_second": 8.582,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 9e-05,
      "loss": 0.5239,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8445333333333334,
      "eval_loss": 0.5255587100982666,
      "eval_runtime": 6.2494,
      "eval_samples_per_second": 600.055,
      "eval_steps_per_second": 9.441,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 18.727033615112305,
      "learning_rate": 9e-05,
      "loss": 0.3429,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8685333333333334,
      "eval_loss": 0.4309305250644684,
      "eval_runtime": 6.3532,
      "eval_samples_per_second": 590.252,
      "eval_steps_per_second": 9.287,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 12.417259216308594,
      "learning_rate": 9e-05,
      "loss": 0.2489,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.5022030472755432,
      "eval_runtime": 6.3216,
      "eval_samples_per_second": 593.208,
      "eval_steps_per_second": 9.333,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.4722766578197479,
      "learning_rate": 9e-05,
      "loss": 0.1951,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4380069077014923,
      "eval_runtime": 6.3792,
      "eval_samples_per_second": 587.85,
      "eval_steps_per_second": 9.249,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.8428453207015991,
      "learning_rate": 9e-05,
      "loss": 0.1474,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8637333333333334,
      "eval_loss": 0.47336435317993164,
      "eval_runtime": 6.4193,
      "eval_samples_per_second": 584.177,
      "eval_steps_per_second": 9.191,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 16.712364196777344,
      "learning_rate": 9e-05,
      "loss": 0.1192,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.4640558660030365,
      "eval_runtime": 6.9906,
      "eval_samples_per_second": 536.432,
      "eval_steps_per_second": 8.44,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}