trainer_state.json

{
  "best_metric": 0.5722666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0981/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": 0.0004287859287859288,
      "loss": 3.066,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.2736,
      "eval_loss": 2.589123010635376,
      "eval_runtime": 7.9177,
      "eval_samples_per_second": 473.625,
      "eval_steps_per_second": 7.452,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 0.00035757185757185754,
      "loss": 2.3659,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3741333333333333,
      "eval_loss": 2.180731773376465,
      "eval_runtime": 6.0812,
      "eval_samples_per_second": 616.656,
      "eval_steps_per_second": 9.702,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": Infinity,
      "learning_rate": 0.0002863577863577864,
      "loss": 1.9149,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4573333333333333,
      "eval_loss": 1.859874963760376,
      "eval_runtime": 5.4946,
      "eval_samples_per_second": 682.493,
      "eval_steps_per_second": 10.738,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 15.396133422851562,
      "learning_rate": 0.00021492921492921493,
      "loss": 1.5148,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4978666666666667,
      "eval_loss": 1.7372242212295532,
      "eval_runtime": 6.2355,
      "eval_samples_per_second": 601.392,
      "eval_steps_per_second": 9.462,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 15.006011009216309,
      "learning_rate": 0.00014350064350064352,
      "loss": 1.0454,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5474666666666667,
      "eval_loss": 1.6076078414916992,
      "eval_runtime": 5.4194,
      "eval_samples_per_second": 691.959,
      "eval_steps_per_second": 10.887,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 11.827376365661621,
      "learning_rate": 7.207207207207208e-05,
      "loss": 0.5162,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5629333333333333,
      "eval_loss": 1.6282234191894531,
      "eval_runtime": 5.4871,
      "eval_samples_per_second": 683.423,
      "eval_steps_per_second": 10.753,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.9008837938308716,
      "learning_rate": 6.435006435006434e-07,
      "loss": 0.1379,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5722666666666667,
      "eval_loss": 1.6859277486801147,
      "eval_runtime": 5.7502,
      "eval_samples_per_second": 652.153,
      "eval_steps_per_second": 10.261,
      "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
}