trainer_state.json

{
  "best_metric": 0.9426666666666667,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0448/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": 4.376876876876877e-05,
      "loss": 0.8584,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9296,
      "eval_loss": 0.260351300239563,
      "eval_runtime": 5.3352,
      "eval_samples_per_second": 702.885,
      "eval_steps_per_second": 11.059,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.38560381531715393,
      "learning_rate": 3.7518768768768766e-05,
      "loss": 0.1416,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.928,
      "eval_loss": 0.238887757062912,
      "eval_runtime": 5.6082,
      "eval_samples_per_second": 668.662,
      "eval_steps_per_second": 10.52,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.07916437834501266,
      "learning_rate": 3.126876876876877e-05,
      "loss": 0.0559,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9370666666666667,
      "eval_loss": 0.22132496535778046,
      "eval_runtime": 5.6499,
      "eval_samples_per_second": 663.727,
      "eval_steps_per_second": 10.443,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.039849355816841125,
      "learning_rate": 2.5018768768768767e-05,
      "loss": 0.0221,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9352,
      "eval_loss": 0.24249306321144104,
      "eval_runtime": 5.3165,
      "eval_samples_per_second": 705.355,
      "eval_steps_per_second": 11.098,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3353838324546814,
      "learning_rate": 1.8768768768768768e-05,
      "loss": 0.0086,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9381333333333334,
      "eval_loss": 0.24046742916107178,
      "eval_runtime": 5.4247,
      "eval_samples_per_second": 691.283,
      "eval_steps_per_second": 10.876,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.022054605185985565,
      "learning_rate": 1.2518768768768768e-05,
      "loss": 0.004,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9405333333333333,
      "eval_loss": 0.24616019427776337,
      "eval_runtime": 5.7311,
      "eval_samples_per_second": 654.328,
      "eval_steps_per_second": 10.295,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.10732956230640411,
      "learning_rate": 6.268768768768769e-06,
      "loss": 0.0025,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9424,
      "eval_loss": 0.24867139756679535,
      "eval_runtime": 5.174,
      "eval_samples_per_second": 724.774,
      "eval_steps_per_second": 11.403,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.11961457133293152,
      "learning_rate": 1.876876876876877e-08,
      "loss": 0.002,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9426666666666667,
      "eval_loss": 0.24936583638191223,
      "eval_runtime": 5.3204,
      "eval_samples_per_second": 704.83,
      "eval_steps_per_second": 11.089,
      "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
}