trainer_state.json

{
  "best_metric": 0.8898666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0199/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": 116.52154541015625,
      "learning_rate": 8.571428571428571e-05,
      "loss": 2.5949,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7784,
      "eval_loss": 0.7949820160865784,
      "eval_runtime": 7.6155,
      "eval_samples_per_second": 492.417,
      "eval_steps_per_second": 7.747,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 75.07592010498047,
      "learning_rate": 7.142857142857143e-05,
      "loss": 0.588,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.4920229911804199,
      "eval_runtime": 7.6387,
      "eval_samples_per_second": 490.921,
      "eval_steps_per_second": 7.724,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 102.26848602294922,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.3508,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8792,
      "eval_loss": 0.3978904187679291,
      "eval_runtime": 7.6344,
      "eval_samples_per_second": 491.196,
      "eval_steps_per_second": 7.728,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 67.49897766113281,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 0.2289,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8845333333333333,
      "eval_loss": 0.37711140513420105,
      "eval_runtime": 7.645,
      "eval_samples_per_second": 490.517,
      "eval_steps_per_second": 7.717,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 165.37841796875,
      "learning_rate": 2.857142857142857e-05,
      "loss": 0.1659,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8837333333333334,
      "eval_loss": 0.3928772807121277,
      "eval_runtime": 7.6734,
      "eval_samples_per_second": 488.703,
      "eval_steps_per_second": 7.689,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 133.74700927734375,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 0.123,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8898666666666667,
      "eval_loss": 0.3759635388851166,
      "eval_runtime": 7.7152,
      "eval_samples_per_second": 486.052,
      "eval_steps_per_second": 7.647,
      "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
}