trainer_state.json

{
  "best_metric": 0.9424,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0637/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": 12.51185131072998,
      "learning_rate": 2.886495269365963e-05,
      "loss": 1.2721,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9181333333333334,
      "eval_loss": 0.3369801938533783,
      "eval_runtime": 7.3785,
      "eval_samples_per_second": 508.231,
      "eval_steps_per_second": 7.996,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.32116183638572693,
      "learning_rate": 2.5619102456015873e-05,
      "loss": 0.2019,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9312,
      "eval_loss": 0.2428940385580063,
      "eval_runtime": 7.0066,
      "eval_samples_per_second": 535.21,
      "eval_steps_per_second": 8.421,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 9.690787315368652,
      "learning_rate": 2.0756590131847176e-05,
      "loss": 0.0928,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9354666666666667,
      "eval_loss": 0.22921998798847198,
      "eval_runtime": 7.4312,
      "eval_samples_per_second": 504.626,
      "eval_steps_per_second": 7.939,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.056311652064323425,
      "learning_rate": 1.5017689143724237e-05,
      "loss": 0.0411,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9408,
      "eval_loss": 0.23159261047840118,
      "eval_runtime": 7.145,
      "eval_samples_per_second": 524.845,
      "eval_steps_per_second": 8.258,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.2053319215774536,
      "learning_rate": 9.276095143821772e-06,
      "loss": 0.0204,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9386666666666666,
      "eval_loss": 0.23609279096126556,
      "eval_runtime": 7.6845,
      "eval_samples_per_second": 487.994,
      "eval_steps_per_second": 7.678,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.13561122119426727,
      "learning_rate": 4.4059137709457125e-06,
      "loss": 0.0121,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9394666666666667,
      "eval_loss": 0.231462761759758,
      "eval_runtime": 7.6025,
      "eval_samples_per_second": 493.258,
      "eval_steps_per_second": 7.761,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.3936139643192291,
      "learning_rate": 1.1485859908123557e-06,
      "loss": 0.0086,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.23656097054481506,
      "eval_runtime": 7.2477,
      "eval_samples_per_second": 517.404,
      "eval_steps_per_second": 8.14,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.13843934237957,
      "learning_rate": 1.04301971493026e-11,
      "loss": 0.0077,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9424,
      "eval_loss": 0.22819308936595917,
      "eval_runtime": 7.3572,
      "eval_samples_per_second": 509.708,
      "eval_steps_per_second": 8.019,
      "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
}