trainer_state.json

{
  "best_metric": 0.8973333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0451/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": 171.1582794189453,
      "learning_rate": 9.698463103929542e-05,
      "loss": 2.6554,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7605333333333333,
      "eval_loss": 0.8721904754638672,
      "eval_runtime": 5.0973,
      "eval_samples_per_second": 735.681,
      "eval_steps_per_second": 11.575,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 67.36498260498047,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.6225,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.48302987217903137,
      "eval_runtime": 4.8547,
      "eval_samples_per_second": 772.444,
      "eval_steps_per_second": 12.153,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 72.43411254882812,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.3564,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.39784252643585205,
      "eval_runtime": 4.9814,
      "eval_samples_per_second": 752.797,
      "eval_steps_per_second": 11.844,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 100.27137756347656,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.2292,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8832,
      "eval_loss": 0.39469626545906067,
      "eval_runtime": 4.8606,
      "eval_samples_per_second": 771.514,
      "eval_steps_per_second": 12.138,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 120.61796569824219,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.1537,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8957333333333334,
      "eval_loss": 0.3735611140727997,
      "eval_runtime": 4.8564,
      "eval_samples_per_second": 772.177,
      "eval_steps_per_second": 12.149,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 63.76533508300781,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.1072,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8973333333333333,
      "eval_loss": 0.35771444439888,
      "eval_runtime": 4.9008,
      "eval_samples_per_second": 765.184,
      "eval_steps_per_second": 12.039,
      "step": 1998
    }
  ],
  "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": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}