trainer_state.json

{
  "best_metric": 0.8336,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0138/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": 170.1170196533203,
      "learning_rate": 4.375e-05,
      "loss": 3.7295,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4176,
      "eval_loss": 3.2697651386260986,
      "eval_runtime": 15.2253,
      "eval_samples_per_second": 246.301,
      "eval_steps_per_second": 3.875,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 244.08221435546875,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 2.1065,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6824,
      "eval_loss": 1.2888131141662598,
      "eval_runtime": 15.2954,
      "eval_samples_per_second": 245.172,
      "eval_steps_per_second": 3.857,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 193.1626739501953,
      "learning_rate": 3.125e-05,
      "loss": 1.0711,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7618666666666667,
      "eval_loss": 0.8568252325057983,
      "eval_runtime": 15.2329,
      "eval_samples_per_second": 246.178,
      "eval_steps_per_second": 3.873,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 91.25299072265625,
      "learning_rate": 2.5e-05,
      "loss": 0.7981,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7986666666666666,
      "eval_loss": 0.693415641784668,
      "eval_runtime": 15.23,
      "eval_samples_per_second": 246.225,
      "eval_steps_per_second": 3.874,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 119.0737075805664,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.6736,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8098666666666666,
      "eval_loss": 0.6269340515136719,
      "eval_runtime": 15.2027,
      "eval_samples_per_second": 246.667,
      "eval_steps_per_second": 3.881,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 151.48049926757812,
      "learning_rate": 1.25e-05,
      "loss": 0.5959,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8269333333333333,
      "eval_loss": 0.5765226483345032,
      "eval_runtime": 15.1672,
      "eval_samples_per_second": 247.244,
      "eval_steps_per_second": 3.89,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 125.43296813964844,
      "learning_rate": 6.25e-06,
      "loss": 0.5647,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.828,
      "eval_loss": 0.5699335932731628,
      "eval_runtime": 15.1765,
      "eval_samples_per_second": 247.093,
      "eval_steps_per_second": 3.888,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 183.90872192382812,
      "learning_rate": 0.0,
      "loss": 0.5301,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8336,
      "eval_loss": 0.5400943160057068,
      "eval_runtime": 15.1896,
      "eval_samples_per_second": 246.88,
      "eval_steps_per_second": 3.884,
      "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
}