trainer_state.json

{
  "best_metric": 0.8642666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0490/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": 2.9280781745910645,
      "learning_rate": 7e-05,
      "loss": 1.6891,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7816,
      "eval_loss": 0.7952107787132263,
      "eval_runtime": 11.9906,
      "eval_samples_per_second": 312.745,
      "eval_steps_per_second": 4.921,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 7e-05,
      "loss": 0.5003,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8306666666666667,
      "eval_loss": 0.5694382190704346,
      "eval_runtime": 12.0522,
      "eval_samples_per_second": 311.147,
      "eval_steps_per_second": 4.895,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.2376394271850586,
      "learning_rate": 7e-05,
      "loss": 0.2593,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5550471544265747,
      "eval_runtime": 11.8654,
      "eval_samples_per_second": 316.046,
      "eval_steps_per_second": 4.972,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.21286128461360931,
      "learning_rate": 7e-05,
      "loss": 0.15,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8608,
      "eval_loss": 0.4935701787471771,
      "eval_runtime": 11.7457,
      "eval_samples_per_second": 319.265,
      "eval_steps_per_second": 5.023,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3042870759963989,
      "learning_rate": 7e-05,
      "loss": 0.0965,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8466666666666667,
      "eval_loss": 0.5628167986869812,
      "eval_runtime": 12.8043,
      "eval_samples_per_second": 292.871,
      "eval_steps_per_second": 4.608,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.10382553935050964,
      "learning_rate": 7e-05,
      "loss": 0.0711,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.5881074666976929,
      "eval_runtime": 11.8458,
      "eval_samples_per_second": 316.567,
      "eval_steps_per_second": 4.981,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.038670603185892105,
      "learning_rate": 7e-05,
      "loss": 0.0618,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8642666666666666,
      "eval_loss": 0.5586597919464111,
      "eval_runtime": 11.959,
      "eval_samples_per_second": 313.571,
      "eval_steps_per_second": 4.934,
      "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
}