trainer_state.json

{
  "best_metric": 0.8592,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0525/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 40.849517822265625,
      "learning_rate": 9e-05,
      "loss": 3.1781,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6306666666666667,
      "eval_loss": 1.4338464736938477,
      "eval_runtime": 15.1778,
      "eval_samples_per_second": 247.071,
      "eval_steps_per_second": 3.887,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 74.61392974853516,
      "learning_rate": 9e-05,
      "loss": 0.9439,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7917333333333333,
      "eval_loss": 0.6849919557571411,
      "eval_runtime": 15.323,
      "eval_samples_per_second": 244.73,
      "eval_steps_per_second": 3.85,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 100.45657348632812,
      "learning_rate": 9e-05,
      "loss": 0.5664,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8242666666666667,
      "eval_loss": 0.5783458352088928,
      "eval_runtime": 14.7716,
      "eval_samples_per_second": 253.865,
      "eval_steps_per_second": 3.994,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 51.926326751708984,
      "learning_rate": 9e-05,
      "loss": 0.427,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5299595594406128,
      "eval_runtime": 15.1894,
      "eval_samples_per_second": 246.882,
      "eval_steps_per_second": 3.884,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 32.73386001586914,
      "learning_rate": 9e-05,
      "loss": 0.3247,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8477333333333333,
      "eval_loss": 0.4970811605453491,
      "eval_runtime": 15.1041,
      "eval_samples_per_second": 248.277,
      "eval_steps_per_second": 3.906,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 149.561279296875,
      "learning_rate": 9e-05,
      "loss": 0.2721,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.47298604249954224,
      "eval_runtime": 15.0817,
      "eval_samples_per_second": 248.646,
      "eval_steps_per_second": 3.912,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 120.08071899414062,
      "learning_rate": 9e-05,
      "loss": 0.2093,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.48994874954223633,
      "eval_runtime": 15.1366,
      "eval_samples_per_second": 247.743,
      "eval_steps_per_second": 3.898,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 61.044944763183594,
      "learning_rate": 9e-05,
      "loss": 0.1707,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8536,
      "eval_loss": 0.5194127559661865,
      "eval_runtime": 15.0406,
      "eval_samples_per_second": 249.324,
      "eval_steps_per_second": 3.923,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 70.00299835205078,
      "learning_rate": 9e-05,
      "loss": 0.1395,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.5127668380737305,
      "eval_runtime": 15.1779,
      "eval_samples_per_second": 247.069,
      "eval_steps_per_second": 3.887,
      "step": 2997
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}