trainer_state.json

{
  "best_metric": 0.9021333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0592/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": 5e-05,
      "loss": 1.762,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7954666666666667,
      "eval_loss": 0.7719430327415466,
      "eval_runtime": 7.0881,
      "eval_samples_per_second": 529.059,
      "eval_steps_per_second": 8.324,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 5.1961541175842285,
      "learning_rate": 5e-05,
      "loss": 0.5607,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.4909805655479431,
      "eval_runtime": 7.3331,
      "eval_samples_per_second": 511.382,
      "eval_steps_per_second": 8.046,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 2.3196747303009033,
      "learning_rate": 5e-05,
      "loss": 0.3605,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.4488397240638733,
      "eval_runtime": 6.8401,
      "eval_samples_per_second": 548.24,
      "eval_steps_per_second": 8.626,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 30.590900421142578,
      "learning_rate": 5e-05,
      "loss": 0.2481,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4282858669757843,
      "eval_runtime": 7.4548,
      "eval_samples_per_second": 503.031,
      "eval_steps_per_second": 7.914,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 6.736176013946533,
      "learning_rate": 5e-05,
      "loss": 0.1863,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.884,
      "eval_loss": 0.4091653823852539,
      "eval_runtime": 6.9059,
      "eval_samples_per_second": 543.015,
      "eval_steps_per_second": 8.543,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 5.18507719039917,
      "learning_rate": 5e-05,
      "loss": 0.1383,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.41286763548851013,
      "eval_runtime": 7.1665,
      "eval_samples_per_second": 523.266,
      "eval_steps_per_second": 8.233,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 11.045472145080566,
      "learning_rate": 5e-05,
      "loss": 0.1131,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9021333333333333,
      "eval_loss": 0.3829288184642792,
      "eval_runtime": 6.9221,
      "eval_samples_per_second": 541.74,
      "eval_steps_per_second": 8.523,
      "step": 2331
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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
}