trainer_state.json

{
  "best_metric": 0.8592,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0072/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": 114.52349090576172,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 3.5965,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5506666666666666,
      "eval_loss": 2.8107409477233887,
      "eval_runtime": 7.3573,
      "eval_samples_per_second": 509.699,
      "eval_steps_per_second": 8.019,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 232.50970458984375,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.5908,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7754666666666666,
      "eval_loss": 0.8627175688743591,
      "eval_runtime": 7.4205,
      "eval_samples_per_second": 505.357,
      "eval_steps_per_second": 7.951,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 212.16748046875,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.7337,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8277333333333333,
      "eval_loss": 0.604835033416748,
      "eval_runtime": 7.3734,
      "eval_samples_per_second": 508.588,
      "eval_steps_per_second": 8.002,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 69.53297424316406,
      "learning_rate": 2.5e-05,
      "loss": 0.5241,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.4963006377220154,
      "eval_runtime": 7.4148,
      "eval_samples_per_second": 505.742,
      "eval_steps_per_second": 7.957,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 90.21280670166016,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.4434,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8570666666666666,
      "eval_loss": 0.456428200006485,
      "eval_runtime": 7.5548,
      "eval_samples_per_second": 496.373,
      "eval_steps_per_second": 7.81,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 115.95838165283203,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.3804,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8568,
      "eval_loss": 0.45551764965057373,
      "eval_runtime": 7.3748,
      "eval_samples_per_second": 508.487,
      "eval_steps_per_second": 8.0,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 149.44200134277344,
      "learning_rate": 1.9030116872178316e-06,
      "loss": 0.3646,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.4566854238510132,
      "eval_runtime": 7.4136,
      "eval_samples_per_second": 505.829,
      "eval_steps_per_second": 7.958,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 77.72151184082031,
      "learning_rate": 0.0,
      "loss": 0.3543,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8592,
      "eval_loss": 0.44104665517807007,
      "eval_runtime": 7.4491,
      "eval_samples_per_second": 503.414,
      "eval_steps_per_second": 7.92,
      "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
}