trainer_state.json

{
  "best_metric": 0.8589333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/facebook_vit-mae-base/model_idx_0744/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": 0.0002887169311775794,
      "loss": 2.1573,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6818666666666666,
      "eval_loss": 1.077835202217102,
      "eval_runtime": 11.2063,
      "eval_samples_per_second": 334.635,
      "eval_steps_per_second": 5.265,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.3217036724090576,
      "learning_rate": 0.00025631588426322575,
      "loss": 0.7962,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7368,
      "eval_loss": 0.8902368545532227,
      "eval_runtime": 11.1569,
      "eval_samples_per_second": 336.115,
      "eval_steps_per_second": 5.288,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 10.089679718017578,
      "learning_rate": 0.0002077292077256867,
      "loss": 0.4765,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8005333333333333,
      "eval_loss": 0.6636611223220825,
      "eval_runtime": 11.3998,
      "eval_samples_per_second": 328.952,
      "eval_steps_per_second": 5.176,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 29.784317016601562,
      "learning_rate": 0.00015035378262848306,
      "loss": 0.2614,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8226666666666667,
      "eval_loss": 0.6182366013526917,
      "eval_runtime": 11.6837,
      "eval_samples_per_second": 320.961,
      "eval_steps_per_second": 5.05,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.6210642457008362,
      "learning_rate": 9.29244973331404e-05,
      "loss": 0.1255,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8282666666666667,
      "eval_loss": 0.6030340790748596,
      "eval_runtime": 11.2822,
      "eval_samples_per_second": 332.382,
      "eval_steps_per_second": 5.229,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 8.319759368896484,
      "learning_rate": 4.4184439928107014e-05,
      "loss": 0.0463,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5575776696205139,
      "eval_runtime": 11.5862,
      "eval_samples_per_second": 323.661,
      "eval_steps_per_second": 5.092,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.014490477740764618,
      "learning_rate": 1.1553842323578666e-05,
      "loss": 0.0163,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8589333333333333,
      "eval_loss": 0.5459588766098022,
      "eval_runtime": 11.173,
      "eval_samples_per_second": 335.631,
      "eval_steps_per_second": 5.281,
      "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
}