trainer_state.json

{
  "best_metric": 0.9464,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0507/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": 2.667667667667668e-05,
      "loss": 1.1362,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9322666666666667,
      "eval_loss": 0.269392728805542,
      "eval_runtime": 13.2149,
      "eval_samples_per_second": 283.771,
      "eval_steps_per_second": 4.465,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.6958529949188232,
      "learning_rate": 2.3343343343343343e-05,
      "loss": 0.1383,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9402666666666667,
      "eval_loss": 0.2050878405570984,
      "eval_runtime": 12.9951,
      "eval_samples_per_second": 288.571,
      "eval_steps_per_second": 4.54,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.3550627827644348,
      "learning_rate": 2.001001001001001e-05,
      "loss": 0.0498,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9405333333333333,
      "eval_loss": 0.20566989481449127,
      "eval_runtime": 12.1271,
      "eval_samples_per_second": 309.226,
      "eval_steps_per_second": 4.865,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.13186924159526825,
      "learning_rate": 1.6676676676676678e-05,
      "loss": 0.0182,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9421333333333334,
      "eval_loss": 0.20005419850349426,
      "eval_runtime": 12.0676,
      "eval_samples_per_second": 310.749,
      "eval_steps_per_second": 4.889,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 1.4917566776275635,
      "learning_rate": 1.3343343343343344e-05,
      "loss": 0.0085,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9450666666666667,
      "eval_loss": 0.1979070007801056,
      "eval_runtime": 12.1026,
      "eval_samples_per_second": 309.851,
      "eval_steps_per_second": 4.875,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.07210291177034378,
      "learning_rate": 1.001001001001001e-05,
      "loss": 0.0052,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9456,
      "eval_loss": 0.19830970466136932,
      "eval_runtime": 12.1714,
      "eval_samples_per_second": 308.1,
      "eval_steps_per_second": 4.847,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.03020659275352955,
      "learning_rate": 6.676676676676677e-06,
      "loss": 0.0037,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9442666666666667,
      "eval_loss": 0.2010853886604309,
      "eval_runtime": 12.0577,
      "eval_samples_per_second": 311.003,
      "eval_steps_per_second": 4.893,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.08498670160770416,
      "learning_rate": 3.3433433433433436e-06,
      "loss": 0.0026,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9464,
      "eval_loss": 0.20677410066127777,
      "eval_runtime": 12.0886,
      "eval_samples_per_second": 310.21,
      "eval_steps_per_second": 4.881,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 1.317930564980736e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}