trainer_state.json

{
  "best_metric": 0.9578666666666666,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0377/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": 2.403313398361206,
      "learning_rate": 8.571428571428571e-05,
      "loss": 0.573,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9304,
      "eval_loss": 0.23466019332408905,
      "eval_runtime": 7.1869,
      "eval_samples_per_second": 521.786,
      "eval_steps_per_second": 8.209,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.05975443497300148,
      "learning_rate": 7.142857142857143e-05,
      "loss": 0.1223,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9368,
      "eval_loss": 0.21442921459674835,
      "eval_runtime": 7.109,
      "eval_samples_per_second": 527.501,
      "eval_steps_per_second": 8.299,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.06025068089365959,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.0542,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9424,
      "eval_loss": 0.2097039371728897,
      "eval_runtime": 7.0299,
      "eval_samples_per_second": 533.439,
      "eval_steps_per_second": 8.393,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.01321440190076828,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.0254,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9474666666666667,
      "eval_loss": 0.20420381426811218,
      "eval_runtime": 6.6654,
      "eval_samples_per_second": 562.61,
      "eval_steps_per_second": 8.852,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.01703156717121601,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.0109,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9466666666666667,
      "eval_loss": 0.19640003144741058,
      "eval_runtime": 7.0995,
      "eval_samples_per_second": 528.208,
      "eval_steps_per_second": 8.31,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.03392144292593002,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.0045,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9578666666666666,
      "eval_loss": 0.18745481967926025,
      "eval_runtime": 7.1403,
      "eval_samples_per_second": 525.187,
      "eval_steps_per_second": 8.263,
      "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
}