trainer_state.json

{
  "best_metric": 0.9405333333333333,
  "best_model_checkpoint": "./vit_finetuned_models_dataset/CIFAR100/50_from_100/google_vit-base-patch16-224/model_idx_0475/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": 0.6405339241027832,
      "learning_rate": 3e-05,
      "loss": 1.1403,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.928,
      "eval_loss": 0.2934419810771942,
      "eval_runtime": 5.6847,
      "eval_samples_per_second": 659.669,
      "eval_steps_per_second": 10.379,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": Infinity,
      "learning_rate": 3e-05,
      "loss": 0.1711,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.924,
      "eval_loss": 0.2609027326107025,
      "eval_runtime": 5.6033,
      "eval_samples_per_second": 669.249,
      "eval_steps_per_second": 10.53,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 46.53877258300781,
      "learning_rate": 3e-05,
      "loss": 0.086,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9389333333333333,
      "eval_loss": 0.22307446599006653,
      "eval_runtime": 5.7208,
      "eval_samples_per_second": 655.5,
      "eval_steps_per_second": 10.313,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.1533428579568863,
      "learning_rate": 3e-05,
      "loss": 0.0402,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9357333333333333,
      "eval_loss": 0.2366916537284851,
      "eval_runtime": 5.6693,
      "eval_samples_per_second": 661.462,
      "eval_steps_per_second": 10.407,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.12214963883161545,
      "learning_rate": 3e-05,
      "loss": 0.0245,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.94,
      "eval_loss": 0.2501796782016754,
      "eval_runtime": 5.6652,
      "eval_samples_per_second": 661.931,
      "eval_steps_per_second": 10.414,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 0.23254963755607605,
      "learning_rate": 3e-05,
      "loss": 0.016,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9405333333333333,
      "eval_loss": 0.2678111493587494,
      "eval_runtime": 6.5047,
      "eval_samples_per_second": 576.502,
      "eval_steps_per_second": 9.07,
      "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
}