trainer_state.json

{
  "best_metric": 0.8792,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0400/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": 235.71817016601562,
      "learning_rate": 3e-05,
      "loss": 3.8304,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.44666666666666666,
      "eval_loss": 3.655517339706421,
      "eval_runtime": 5.1815,
      "eval_samples_per_second": 723.73,
      "eval_steps_per_second": 11.387,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 304.77337646484375,
      "learning_rate": 3e-05,
      "loss": 2.895,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6789333333333334,
      "eval_loss": 1.777080774307251,
      "eval_runtime": 5.301,
      "eval_samples_per_second": 707.417,
      "eval_steps_per_second": 11.13,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 145.64024353027344,
      "learning_rate": 3e-05,
      "loss": 1.1523,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8018666666666666,
      "eval_loss": 0.7707297801971436,
      "eval_runtime": 5.1462,
      "eval_samples_per_second": 728.699,
      "eval_steps_per_second": 11.465,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 307.16998291015625,
      "learning_rate": 3e-05,
      "loss": 0.6834,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8338666666666666,
      "eval_loss": 0.5799203515052795,
      "eval_runtime": 5.1334,
      "eval_samples_per_second": 730.516,
      "eval_steps_per_second": 11.493,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 316.3178405761719,
      "learning_rate": 3e-05,
      "loss": 0.5185,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8597333333333333,
      "eval_loss": 0.4851472079753876,
      "eval_runtime": 5.0904,
      "eval_samples_per_second": 736.674,
      "eval_steps_per_second": 11.59,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 200.60751342773438,
      "learning_rate": 3e-05,
      "loss": 0.4245,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8629333333333333,
      "eval_loss": 0.4591628909111023,
      "eval_runtime": 5.2688,
      "eval_samples_per_second": 711.739,
      "eval_steps_per_second": 11.198,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 148.9307098388672,
      "learning_rate": 3e-05,
      "loss": 0.3696,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.42555832862854004,
      "eval_runtime": 5.117,
      "eval_samples_per_second": 732.854,
      "eval_steps_per_second": 11.53,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 127.84893035888672,
      "learning_rate": 3e-05,
      "loss": 0.3077,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.3858112394809723,
      "eval_runtime": 5.2352,
      "eval_samples_per_second": 716.308,
      "eval_steps_per_second": 11.27,
      "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
}