trainer_state.json

{
  "best_metric": 0.9461333333333334,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0201/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.9638573527336121,
      "learning_rate": 3e-05,
      "loss": 1.2022,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9285333333333333,
      "eval_loss": 0.3046490550041199,
      "eval_runtime": 8.9246,
      "eval_samples_per_second": 420.187,
      "eval_steps_per_second": 6.611,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.934065818786621,
      "learning_rate": 3e-05,
      "loss": 0.1738,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9386666666666666,
      "eval_loss": 0.21978728473186493,
      "eval_runtime": 7.2048,
      "eval_samples_per_second": 520.489,
      "eval_steps_per_second": 8.189,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.4749391973018646,
      "learning_rate": 3e-05,
      "loss": 0.0711,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9378666666666666,
      "eval_loss": 0.21729663014411926,
      "eval_runtime": 6.8978,
      "eval_samples_per_second": 543.648,
      "eval_steps_per_second": 8.553,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.036638252437114716,
      "learning_rate": 3e-05,
      "loss": 0.0329,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.944,
      "eval_loss": 0.2027207463979721,
      "eval_runtime": 6.8958,
      "eval_samples_per_second": 543.813,
      "eval_steps_per_second": 8.556,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.02701728232204914,
      "learning_rate": 3e-05,
      "loss": 0.017,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9405333333333333,
      "eval_loss": 0.21887874603271484,
      "eval_runtime": 6.9194,
      "eval_samples_per_second": 541.958,
      "eval_steps_per_second": 8.527,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.679091215133667,
      "learning_rate": 3e-05,
      "loss": 0.0089,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9461333333333334,
      "eval_loss": 0.22340188920497894,
      "eval_runtime": 6.9192,
      "eval_samples_per_second": 541.97,
      "eval_steps_per_second": 8.527,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 9.88447923735552e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}