trainer_state.json

{
  "best_metric": 0.8992,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0996/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "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": 0.0004810825448943271,
      "loss": 0.8627,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7858666666666667,
      "eval_loss": 0.7205981612205505,
      "eval_runtime": 6.2188,
      "eval_samples_per_second": 603.015,
      "eval_steps_per_second": 9.487,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.346010208129883,
      "learning_rate": 0.0004269850409335979,
      "loss": 0.5042,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7885333333333333,
      "eval_loss": 0.7182707190513611,
      "eval_runtime": 6.1559,
      "eval_samples_per_second": 609.171,
      "eval_steps_per_second": 9.584,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.945357322692871,
      "learning_rate": 0.0003459431688641196,
      "loss": 0.3342,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8229333333333333,
      "eval_loss": 0.6171352863311768,
      "eval_runtime": 6.1406,
      "eval_samples_per_second": 610.694,
      "eval_steps_per_second": 9.608,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.06508439779281616,
      "learning_rate": 0.00025029481906207064,
      "loss": 0.1951,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8602666666666666,
      "eval_loss": 0.494656503200531,
      "eval_runtime": 6.1623,
      "eval_samples_per_second": 608.54,
      "eval_steps_per_second": 9.574,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.828187108039856,
      "learning_rate": 0.00015460158573036287,
      "loss": 0.1015,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8666666666666667,
      "eval_loss": 0.5341471433639526,
      "eval_runtime": 6.1641,
      "eval_samples_per_second": 608.366,
      "eval_steps_per_second": 9.572,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.09762649238109589,
      "learning_rate": 7.343189618242854e-05,
      "loss": 0.0379,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8890666666666667,
      "eval_loss": 0.4427512288093567,
      "eval_runtime": 6.0869,
      "eval_samples_per_second": 616.081,
      "eval_steps_per_second": 9.693,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.02568543143570423,
      "learning_rate": 1.9143099846872597e-05,
      "loss": 0.0118,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8978666666666667,
      "eval_loss": 0.41292688250541687,
      "eval_runtime": 6.1805,
      "eval_samples_per_second": 606.745,
      "eval_steps_per_second": 9.546,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.034303903579711914,
      "learning_rate": 1.7383661915504333e-10,
      "loss": 0.0057,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.4011141359806061,
      "eval_runtime": 6.1243,
      "eval_samples_per_second": 612.319,
      "eval_steps_per_second": 9.634,
      "step": 2664
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}