trainer_state.json

{
  "best_metric": 0.9104,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0145/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": 0.0004376876876876877,
      "loss": 0.8116,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8250666666666666,
      "eval_loss": 0.6092773675918579,
      "eval_runtime": 7.4536,
      "eval_samples_per_second": 503.113,
      "eval_steps_per_second": 7.916,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 34.006263732910156,
      "learning_rate": 0.0003751876876876877,
      "loss": 0.4446,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8088,
      "eval_loss": 0.6750194430351257,
      "eval_runtime": 7.0215,
      "eval_samples_per_second": 534.077,
      "eval_steps_per_second": 8.403,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.088271617889404,
      "learning_rate": 0.0003126876876876877,
      "loss": 0.2791,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.852,
      "eval_loss": 0.5162176489830017,
      "eval_runtime": 6.3932,
      "eval_samples_per_second": 586.561,
      "eval_steps_per_second": 9.229,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 44.37331008911133,
      "learning_rate": 0.0002501876876876877,
      "loss": 0.1783,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8637333333333334,
      "eval_loss": 0.5190310478210449,
      "eval_runtime": 6.5285,
      "eval_samples_per_second": 574.402,
      "eval_steps_per_second": 9.037,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.0164837297052145,
      "learning_rate": 0.00018768768768768769,
      "loss": 0.0978,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8765333333333334,
      "eval_loss": 0.48268938064575195,
      "eval_runtime": 6.2173,
      "eval_samples_per_second": 603.159,
      "eval_steps_per_second": 9.49,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.009621452540159225,
      "learning_rate": 0.00012518768768768768,
      "loss": 0.0475,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8962666666666667,
      "eval_loss": 0.4216897487640381,
      "eval_runtime": 6.2311,
      "eval_samples_per_second": 601.821,
      "eval_steps_per_second": 9.469,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 0.7187299132347107,
      "learning_rate": 6.26876876876877e-05,
      "loss": 0.0152,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.3846963942050934,
      "eval_runtime": 6.767,
      "eval_samples_per_second": 554.163,
      "eval_steps_per_second": 8.719,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 0.01042836345732212,
      "learning_rate": 1.876876876876877e-07,
      "loss": 0.0064,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9104,
      "eval_loss": 0.3970121145248413,
      "eval_runtime": 6.3064,
      "eval_samples_per_second": 594.63,
      "eval_steps_per_second": 9.356,
      "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
}