trainer_state.json

{
  "best_metric": 0.8730666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0888/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 115.08124542236328,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.7413,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7061333333333333,
      "eval_loss": 1.030971646308899,
      "eval_runtime": 7.5578,
      "eval_samples_per_second": 496.173,
      "eval_steps_per_second": 7.806,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 45.78676223754883,
      "learning_rate": 8.535533905932738e-05,
      "loss": 0.7569,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.808,
      "eval_loss": 0.6007307767868042,
      "eval_runtime": 7.6133,
      "eval_samples_per_second": 492.561,
      "eval_steps_per_second": 7.75,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 62.99378204345703,
      "learning_rate": 6.91341716182545e-05,
      "loss": 0.4928,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.505699634552002,
      "eval_runtime": 7.5145,
      "eval_samples_per_second": 499.034,
      "eval_steps_per_second": 7.851,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 52.81418228149414,
      "learning_rate": 5e-05,
      "loss": 0.3573,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8554666666666667,
      "eval_loss": 0.4670276641845703,
      "eval_runtime": 7.4943,
      "eval_samples_per_second": 500.378,
      "eval_steps_per_second": 7.873,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 161.98309326171875,
      "learning_rate": 3.086582838174551e-05,
      "loss": 0.2851,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.43038859963417053,
      "eval_runtime": 7.6947,
      "eval_samples_per_second": 487.348,
      "eval_steps_per_second": 7.668,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 47.56634521484375,
      "learning_rate": 1.4644660940672627e-05,
      "loss": 0.2332,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8658666666666667,
      "eval_loss": 0.42409420013427734,
      "eval_runtime": 7.6954,
      "eval_samples_per_second": 487.305,
      "eval_steps_per_second": 7.667,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 124.73308563232422,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 0.2097,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8730666666666667,
      "eval_loss": 0.41754284501075745,
      "eval_runtime": 8.1201,
      "eval_samples_per_second": 461.818,
      "eval_steps_per_second": 7.266,
      "step": 2331
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}