trainer_state.json

{
  "best_metric": 0.8930666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0911/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 97.49637603759766,
      "learning_rate": 6.653391037658466e-05,
      "loss": 3.2692,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7085333333333333,
      "eval_loss": 1.4550378322601318,
      "eval_runtime": 8.1629,
      "eval_samples_per_second": 459.398,
      "eval_steps_per_second": 7.228,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 81.96347045898438,
      "learning_rate": 5.682214306505567e-05,
      "loss": 0.8148,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8485333333333334,
      "eval_loss": 0.5207698941230774,
      "eval_runtime": 8.2193,
      "eval_samples_per_second": 456.241,
      "eval_steps_per_second": 7.178,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 219.00494384765625,
      "learning_rate": 4.2788232688471e-05,
      "loss": 0.4298,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8752,
      "eval_loss": 0.4215531051158905,
      "eval_runtime": 8.5551,
      "eval_samples_per_second": 438.334,
      "eval_steps_per_second": 6.896,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 46.776397705078125,
      "learning_rate": 2.7211767311528996e-05,
      "loss": 0.2928,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8864,
      "eval_loss": 0.3718867599964142,
      "eval_runtime": 8.3129,
      "eval_samples_per_second": 451.109,
      "eval_steps_per_second": 7.097,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 201.11160278320312,
      "learning_rate": 1.3177856934944328e-05,
      "loss": 0.2268,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8906666666666667,
      "eval_loss": 0.3597862422466278,
      "eval_runtime": 8.3414,
      "eval_samples_per_second": 449.562,
      "eval_steps_per_second": 7.073,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 111.83905029296875,
      "learning_rate": 3.4660896234153336e-06,
      "loss": 0.1875,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8930666666666667,
      "eval_loss": 0.35456985235214233,
      "eval_runtime": 8.2909,
      "eval_samples_per_second": 452.302,
      "eval_steps_per_second": 7.116,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}