trainer_state.json

{
  "best_metric": 0.8917333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0810/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": 9.621650897886543e-05,
      "loss": 1.4463,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7952,
      "eval_loss": 0.6925199031829834,
      "eval_runtime": 7.3222,
      "eval_samples_per_second": 512.139,
      "eval_steps_per_second": 8.058,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 8.543862808774192e-05,
      "loss": 0.4378,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8506666666666667,
      "eval_loss": 0.49212077260017395,
      "eval_runtime": 5.9138,
      "eval_samples_per_second": 634.108,
      "eval_steps_per_second": 9.977,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.2665881812572479,
      "learning_rate": 6.924306924189557e-05,
      "loss": 0.2207,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.4650355279445648,
      "eval_runtime": 5.7242,
      "eval_samples_per_second": 655.11,
      "eval_steps_per_second": 10.307,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.17952285706996918,
      "learning_rate": 5.011792754282769e-05,
      "loss": 0.1088,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8738666666666667,
      "eval_loss": 0.4514040946960449,
      "eval_runtime": 5.6888,
      "eval_samples_per_second": 659.194,
      "eval_steps_per_second": 10.371,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.2325646877288818,
      "learning_rate": 3.0974832444380136e-05,
      "loss": 0.0486,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.4529297947883606,
      "eval_runtime": 5.6646,
      "eval_samples_per_second": 662.004,
      "eval_steps_per_second": 10.416,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.023798868060112,
      "learning_rate": 1.472814664270234e-05,
      "loss": 0.016,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.43425291776657104,
      "eval_runtime": 5.8212,
      "eval_samples_per_second": 644.201,
      "eval_steps_per_second": 10.135,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.03099374659359455,
      "learning_rate": 3.851280774526222e-06,
      "loss": 0.0054,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8917333333333334,
      "eval_loss": 0.4305557310581207,
      "eval_runtime": 5.7902,
      "eval_samples_per_second": 647.647,
      "eval_steps_per_second": 10.19,
      "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
}