trainer_state.json

{
  "best_metric": 0.8816,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0033/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": 79.31631469726562,
      "learning_rate": 7.875e-05,
      "loss": 2.9536,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6773333333333333,
      "eval_loss": 1.2627865076065063,
      "eval_runtime": 15.5864,
      "eval_samples_per_second": 240.594,
      "eval_steps_per_second": 3.785,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 77.8153305053711,
      "learning_rate": 6.75e-05,
      "loss": 0.8661,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8226666666666667,
      "eval_loss": 0.6220800876617432,
      "eval_runtime": 14.3843,
      "eval_samples_per_second": 260.701,
      "eval_steps_per_second": 4.102,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 57.634037017822266,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 0.547,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.832,
      "eval_loss": 0.5607235431671143,
      "eval_runtime": 14.4475,
      "eval_samples_per_second": 259.561,
      "eval_steps_per_second": 4.084,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 212.1357879638672,
      "learning_rate": 4.5e-05,
      "loss": 0.4446,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8618666666666667,
      "eval_loss": 0.46395137906074524,
      "eval_runtime": 14.4552,
      "eval_samples_per_second": 259.422,
      "eval_steps_per_second": 4.082,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 78.74669647216797,
      "learning_rate": 3.375e-05,
      "loss": 0.3763,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.4346161484718323,
      "eval_runtime": 14.4043,
      "eval_samples_per_second": 260.338,
      "eval_steps_per_second": 4.096,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 85.58202362060547,
      "learning_rate": 2.25e-05,
      "loss": 0.3214,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.43199029564857483,
      "eval_runtime": 14.3337,
      "eval_samples_per_second": 261.622,
      "eval_steps_per_second": 4.116,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 154.25253295898438,
      "learning_rate": 1.125e-05,
      "loss": 0.3023,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.4142453372478485,
      "eval_runtime": 14.2715,
      "eval_samples_per_second": 262.761,
      "eval_steps_per_second": 4.134,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 297.16326904296875,
      "learning_rate": 0.0,
      "loss": 0.2695,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8816,
      "eval_loss": 0.39373475313186646,
      "eval_runtime": 14.3338,
      "eval_samples_per_second": 261.62,
      "eval_steps_per_second": 4.116,
      "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}