trainer_state.json

{
  "best_metric": 0.8672,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0637/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": 144.82443237304688,
      "learning_rate": 5e-05,
      "loss": 3.6963,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4709333333333333,
      "eval_loss": 3.096170663833618,
      "eval_runtime": 7.7963,
      "eval_samples_per_second": 480.998,
      "eval_steps_per_second": 7.568,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 531.730224609375,
      "learning_rate": 5e-05,
      "loss": 1.8215,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7194666666666667,
      "eval_loss": 1.0690484046936035,
      "eval_runtime": 7.7205,
      "eval_samples_per_second": 485.72,
      "eval_steps_per_second": 7.642,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 136.83494567871094,
      "learning_rate": 5e-05,
      "loss": 0.9096,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7906666666666666,
      "eval_loss": 0.7313085794448853,
      "eval_runtime": 7.6081,
      "eval_samples_per_second": 492.898,
      "eval_steps_per_second": 7.755,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 48.68620681762695,
      "learning_rate": 5e-05,
      "loss": 0.6456,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8234666666666667,
      "eval_loss": 0.5820752382278442,
      "eval_runtime": 7.7009,
      "eval_samples_per_second": 486.956,
      "eval_steps_per_second": 7.661,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 149.6680450439453,
      "learning_rate": 5e-05,
      "loss": 0.5105,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8453333333333334,
      "eval_loss": 0.5223390460014343,
      "eval_runtime": 7.7243,
      "eval_samples_per_second": 485.479,
      "eval_steps_per_second": 7.638,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 87.81366729736328,
      "learning_rate": 5e-05,
      "loss": 0.421,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8618666666666667,
      "eval_loss": 0.49326109886169434,
      "eval_runtime": 7.7019,
      "eval_samples_per_second": 486.895,
      "eval_steps_per_second": 7.66,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 61.55073165893555,
      "learning_rate": 5e-05,
      "loss": 0.3489,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8605333333333334,
      "eval_loss": 0.45995429158210754,
      "eval_runtime": 7.6547,
      "eval_samples_per_second": 489.892,
      "eval_steps_per_second": 7.708,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 149.32814025878906,
      "learning_rate": 5e-05,
      "loss": 0.2956,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.43900513648986816,
      "eval_runtime": 7.7602,
      "eval_samples_per_second": 483.235,
      "eval_steps_per_second": 7.603,
      "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
}