trainer_state.json

{
  "best_metric": 0.9114666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0842/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": 121.29356384277344,
      "learning_rate": 6.733578363789503e-05,
      "loss": 3.198,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7330666666666666,
      "eval_loss": 1.4489405155181885,
      "eval_runtime": 5.0375,
      "eval_samples_per_second": 744.422,
      "eval_steps_per_second": 11.712,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 105.1215591430664,
      "learning_rate": 5.974873734152915e-05,
      "loss": 0.8681,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.86,
      "eval_loss": 0.5040282607078552,
      "eval_runtime": 4.9949,
      "eval_samples_per_second": 750.764,
      "eval_steps_per_second": 11.812,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 108.02486419677734,
      "learning_rate": 4.839392013277814e-05,
      "loss": 0.424,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.38806888461112976,
      "eval_runtime": 5.052,
      "eval_samples_per_second": 742.282,
      "eval_steps_per_second": 11.679,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 358.32196044921875,
      "learning_rate": 3.5e-05,
      "loss": 0.2889,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8946666666666667,
      "eval_loss": 0.3593757450580597,
      "eval_runtime": 5.1686,
      "eval_samples_per_second": 725.536,
      "eval_steps_per_second": 11.415,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 111.44454193115234,
      "learning_rate": 2.1606079867221858e-05,
      "loss": 0.2018,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9090666666666667,
      "eval_loss": 0.3189147710800171,
      "eval_runtime": 4.997,
      "eval_samples_per_second": 750.455,
      "eval_steps_per_second": 11.807,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 106.11746978759766,
      "learning_rate": 1.0251262658470837e-05,
      "loss": 0.1602,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9114666666666666,
      "eval_loss": 0.3097210228443146,
      "eval_runtime": 5.058,
      "eval_samples_per_second": 741.404,
      "eval_steps_per_second": 11.665,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}