trainer_state.json

{
  "best_metric": 0.8594666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0902/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": 116.01181030273438,
      "learning_rate": 3e-05,
      "loss": 3.8253,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.41013333333333335,
      "eval_loss": 3.6429638862609863,
      "eval_runtime": 7.4937,
      "eval_samples_per_second": 500.422,
      "eval_steps_per_second": 7.873,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 103.37413787841797,
      "learning_rate": 3e-05,
      "loss": 2.9494,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6088,
      "eval_loss": 1.9309278726577759,
      "eval_runtime": 7.4982,
      "eval_samples_per_second": 500.119,
      "eval_steps_per_second": 7.869,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 125.32000732421875,
      "learning_rate": 3e-05,
      "loss": 1.3274,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7565333333333333,
      "eval_loss": 0.9151241183280945,
      "eval_runtime": 7.6268,
      "eval_samples_per_second": 491.688,
      "eval_steps_per_second": 7.736,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 189.87738037109375,
      "learning_rate": 3e-05,
      "loss": 0.8117,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8069333333333333,
      "eval_loss": 0.690345048904419,
      "eval_runtime": 7.5564,
      "eval_samples_per_second": 496.268,
      "eval_steps_per_second": 7.808,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 171.3572540283203,
      "learning_rate": 3e-05,
      "loss": 0.6294,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8325333333333333,
      "eval_loss": 0.581441342830658,
      "eval_runtime": 7.1182,
      "eval_samples_per_second": 526.821,
      "eval_steps_per_second": 8.289,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 129.79537963867188,
      "learning_rate": 3e-05,
      "loss": 0.5181,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8466666666666667,
      "eval_loss": 0.5113083720207214,
      "eval_runtime": 7.1455,
      "eval_samples_per_second": 524.804,
      "eval_steps_per_second": 8.257,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 133.4134063720703,
      "learning_rate": 3e-05,
      "loss": 0.4438,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.48897606134414673,
      "eval_runtime": 9.1664,
      "eval_samples_per_second": 409.101,
      "eval_steps_per_second": 6.437,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 67.1819839477539,
      "learning_rate": 3e-05,
      "loss": 0.3858,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8594666666666667,
      "eval_loss": 0.4570837616920471,
      "eval_runtime": 7.2028,
      "eval_samples_per_second": 520.629,
      "eval_steps_per_second": 8.191,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}