trainer_state.json

{
  "best_metric": 0.9581333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0463/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": 2.824125051498413,
      "learning_rate": 2.8858192987669303e-05,
      "loss": 1.162,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.26659050583839417,
      "eval_runtime": 7.1462,
      "eval_samples_per_second": 524.756,
      "eval_steps_per_second": 8.256,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 1.6864879131317139,
      "learning_rate": 2.5606601717798212e-05,
      "loss": 0.1575,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9448,
      "eval_loss": 0.20292161405086517,
      "eval_runtime": 7.4377,
      "eval_samples_per_second": 504.189,
      "eval_steps_per_second": 7.933,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.08257895708084106,
      "learning_rate": 2.074025148547635e-05,
      "loss": 0.0657,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9525333333333333,
      "eval_loss": 0.1811654269695282,
      "eval_runtime": 6.9404,
      "eval_samples_per_second": 540.312,
      "eval_steps_per_second": 8.501,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.6105066537857056,
      "learning_rate": 1.5e-05,
      "loss": 0.0304,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9530666666666666,
      "eval_loss": 0.17422278225421906,
      "eval_runtime": 7.9198,
      "eval_samples_per_second": 473.497,
      "eval_steps_per_second": 7.45,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.04834752529859543,
      "learning_rate": 9.259748514523654e-06,
      "loss": 0.0142,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9549333333333333,
      "eval_loss": 0.17519499361515045,
      "eval_runtime": 6.9025,
      "eval_samples_per_second": 543.281,
      "eval_steps_per_second": 8.548,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.1390727460384369,
      "learning_rate": 4.393398282201788e-06,
      "loss": 0.0082,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9570666666666666,
      "eval_loss": 0.17499089241027832,
      "eval_runtime": 7.3446,
      "eval_samples_per_second": 510.58,
      "eval_steps_per_second": 8.033,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": NaN,
      "learning_rate": 1.1485859908123557e-06,
      "loss": 0.0064,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9565333333333333,
      "eval_loss": 0.17338763177394867,
      "eval_runtime": 6.9,
      "eval_samples_per_second": 543.475,
      "eval_steps_per_second": 8.551,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.0574193149805069,
      "learning_rate": 1.04301971493026e-11,
      "loss": 0.0061,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9581333333333333,
      "eval_loss": 0.16709984838962555,
      "eval_runtime": 7.0732,
      "eval_samples_per_second": 530.167,
      "eval_steps_per_second": 8.341,
      "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
}