trainer_state.json

{
  "best_metric": 0.9389333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0892/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 8.086507797241211,
      "learning_rate": 5e-05,
      "loss": 0.9028,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9256,
      "eval_loss": 0.26424717903137207,
      "eval_runtime": 13.5384,
      "eval_samples_per_second": 276.991,
      "eval_steps_per_second": 4.358,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8906418085098267,
      "learning_rate": 5e-05,
      "loss": 0.168,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9341333333333334,
      "eval_loss": 0.23150846362113953,
      "eval_runtime": 13.4757,
      "eval_samples_per_second": 278.279,
      "eval_steps_per_second": 4.378,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 1.2329801321029663,
      "learning_rate": 5e-05,
      "loss": 0.0913,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9298666666666666,
      "eval_loss": 0.25696858763694763,
      "eval_runtime": 13.3973,
      "eval_samples_per_second": 279.907,
      "eval_steps_per_second": 4.404,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 12.360350608825684,
      "learning_rate": 5e-05,
      "loss": 0.0521,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9373333333333334,
      "eval_loss": 0.25371044874191284,
      "eval_runtime": 13.1927,
      "eval_samples_per_second": 284.247,
      "eval_steps_per_second": 4.472,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.3150515556335449,
      "learning_rate": 5e-05,
      "loss": 0.0365,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.24791266024112701,
      "eval_runtime": 13.4146,
      "eval_samples_per_second": 279.546,
      "eval_steps_per_second": 4.398,
      "step": 1665
    }
  ],
  "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": 8.2370660311296e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}