trainer_state.json

{
  "best_metric": 0.8797333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0944/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": 4.7538820348083834e-05,
      "loss": 2.1434,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7336,
      "eval_loss": 0.9953354001045227,
      "eval_runtime": 7.2052,
      "eval_samples_per_second": 520.454,
      "eval_steps_per_second": 8.188,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 4.063987381465584e-05,
      "loss": 0.6999,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8248,
      "eval_loss": 0.5908309817314148,
      "eval_runtime": 7.1048,
      "eval_samples_per_second": 527.81,
      "eval_steps_per_second": 8.304,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 3.066152414593445e-05,
      "loss": 0.4113,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8325333333333333,
      "eval_loss": 0.527634859085083,
      "eval_runtime": 7.1864,
      "eval_samples_per_second": 521.816,
      "eval_steps_per_second": 8.21,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 25.60707664489746,
      "learning_rate": 1.95355683908512e-05,
      "loss": 0.2611,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8650666666666667,
      "eval_loss": 0.42205870151519775,
      "eval_runtime": 7.0239,
      "eval_samples_per_second": 533.889,
      "eval_steps_per_second": 8.4,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 29.090328216552734,
      "learning_rate": 9.491910332815573e-06,
      "loss": 0.1663,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.3933185040950775,
      "eval_runtime": 7.5814,
      "eval_samples_per_second": 494.632,
      "eval_steps_per_second": 7.782,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 1.7436997890472412,
      "learning_rate": 2.5198196276040782e-06,
      "loss": 0.1089,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.38708561658859253,
      "eval_runtime": 7.3083,
      "eval_samples_per_second": 513.115,
      "eval_steps_per_second": 8.073,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 13.166748046875,
      "learning_rate": 2.0434646496314548e-10,
      "loss": 0.0864,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.37674108147621155,
      "eval_runtime": 7.2338,
      "eval_samples_per_second": 518.397,
      "eval_steps_per_second": 8.156,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.153189244358144e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}