trainer_state.json

{
  "best_metric": 0.8818666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0650/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": 198.2068328857422,
      "learning_rate": 8.657457896300791e-05,
      "loss": 2.987,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6981333333333334,
      "eval_loss": 1.232658863067627,
      "eval_runtime": 7.5821,
      "eval_samples_per_second": 494.588,
      "eval_steps_per_second": 7.782,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 94.85763549804688,
      "learning_rate": 7.681980515339464e-05,
      "loss": 0.8514,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8213333333333334,
      "eval_loss": 0.6262484788894653,
      "eval_runtime": 7.8911,
      "eval_samples_per_second": 475.22,
      "eval_steps_per_second": 7.477,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 76.2540512084961,
      "learning_rate": 6.222075445642905e-05,
      "loss": 0.5285,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.47353848814964294,
      "eval_runtime": 7.768,
      "eval_samples_per_second": 482.751,
      "eval_steps_per_second": 7.595,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 73.49557495117188,
      "learning_rate": 4.5e-05,
      "loss": 0.4025,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4329341650009155,
      "eval_runtime": 7.6692,
      "eval_samples_per_second": 488.97,
      "eval_steps_per_second": 7.693,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 84.99456024169922,
      "learning_rate": 2.7779245543570963e-05,
      "loss": 0.3254,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.41132044792175293,
      "eval_runtime": 7.4479,
      "eval_samples_per_second": 503.495,
      "eval_steps_per_second": 7.922,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 151.67532348632812,
      "learning_rate": 1.3180194846605365e-05,
      "loss": 0.288,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.43155521154403687,
      "eval_runtime": 7.5313,
      "eval_samples_per_second": 497.92,
      "eval_steps_per_second": 7.834,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 55.87948989868164,
      "learning_rate": 3.425421036992097e-06,
      "loss": 0.2711,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.40467390418052673,
      "eval_runtime": 7.4512,
      "eval_samples_per_second": 503.275,
      "eval_steps_per_second": 7.918,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 92.61456298828125,
      "learning_rate": 0.0,
      "loss": 0.2601,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8818666666666667,
      "eval_loss": 0.3909488618373871,
      "eval_runtime": 7.9057,
      "eval_samples_per_second": 474.343,
      "eval_steps_per_second": 7.463,
      "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
}