trainer_state.json

{
  "best_metric": 0.8848,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0901/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": 123.01940155029297,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.7361,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7218666666666667,
      "eval_loss": 0.9971085786819458,
      "eval_runtime": 8.8684,
      "eval_samples_per_second": 422.85,
      "eval_steps_per_second": 6.653,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 118.78591918945312,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.7895,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8349333333333333,
      "eval_loss": 0.5567114353179932,
      "eval_runtime": 8.6321,
      "eval_samples_per_second": 434.424,
      "eval_steps_per_second": 6.835,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 182.38658142089844,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.5315,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8522666666666666,
      "eval_loss": 0.5013552904129028,
      "eval_runtime": 8.763,
      "eval_samples_per_second": 427.934,
      "eval_steps_per_second": 6.733,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 83.68070220947266,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.4044,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.42109203338623047,
      "eval_runtime": 9.3728,
      "eval_samples_per_second": 400.094,
      "eval_steps_per_second": 6.295,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 86.44019317626953,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.3325,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8682666666666666,
      "eval_loss": 0.42969420552253723,
      "eval_runtime": 8.8449,
      "eval_samples_per_second": 423.975,
      "eval_steps_per_second": 6.671,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 110.50960540771484,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.27,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.4224238693714142,
      "eval_runtime": 8.6727,
      "eval_samples_per_second": 432.389,
      "eval_steps_per_second": 6.803,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 86.52838897705078,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.2429,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.4011433720588684,
      "eval_runtime": 8.8457,
      "eval_samples_per_second": 423.936,
      "eval_steps_per_second": 6.67,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 75.29777526855469,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.2289,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8848,
      "eval_loss": 0.39161813259124756,
      "eval_runtime": 8.7702,
      "eval_samples_per_second": 427.582,
      "eval_steps_per_second": 6.727,
      "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
}