trainer_state.json

{
  "best_metric": 0.6290666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0853/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": 0.0004810825448943271,
      "loss": 3.0032,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.29786666666666667,
      "eval_loss": 2.578808307647705,
      "eval_runtime": 13.7195,
      "eval_samples_per_second": 273.334,
      "eval_steps_per_second": 4.3,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0004271931404387096,
      "loss": 2.2604,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.39413333333333334,
      "eval_loss": 2.142023801803589,
      "eval_runtime": 13.5306,
      "eval_samples_per_second": 277.15,
      "eval_steps_per_second": 4.36,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 8.582889556884766,
      "learning_rate": 0.0003462153462094779,
      "loss": 1.8326,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.47226666666666667,
      "eval_loss": 1.8428372144699097,
      "eval_runtime": 13.521,
      "eval_samples_per_second": 277.346,
      "eval_steps_per_second": 4.364,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 9.992016792297363,
      "learning_rate": 0.00025058963771413845,
      "loss": 1.466,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.516,
      "eval_loss": 1.6780189275741577,
      "eval_runtime": 13.5247,
      "eval_samples_per_second": 277.271,
      "eval_steps_per_second": 4.362,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 5.6153459548950195,
      "learning_rate": 0.00015487416222190068,
      "loss": 1.0774,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5618666666666666,
      "eval_loss": 1.5338225364685059,
      "eval_runtime": 13.7613,
      "eval_samples_per_second": 272.504,
      "eval_steps_per_second": 4.287,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.562627375125885,
      "learning_rate": 7.36407332135117e-05,
      "loss": 0.6838,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6133333333333333,
      "eval_loss": 1.4478487968444824,
      "eval_runtime": 13.9931,
      "eval_samples_per_second": 267.99,
      "eval_steps_per_second": 4.216,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 16.792430877685547,
      "learning_rate": 1.9256403872631112e-05,
      "loss": 0.3819,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6290666666666667,
      "eval_loss": 1.4039924144744873,
      "eval_runtime": 13.5111,
      "eval_samples_per_second": 277.549,
      "eval_steps_per_second": 4.367,
      "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
}