trainer_state.json

{
  "best_metric": 0.8533333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0765/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 100.27824401855469,
      "learning_rate": 7.2e-05,
      "loss": 3.1045,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.684,
      "eval_loss": 1.3956953287124634,
      "eval_runtime": 5.543,
      "eval_samples_per_second": 676.524,
      "eval_steps_per_second": 10.644,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 85.8294906616211,
      "learning_rate": 5.4000000000000005e-05,
      "loss": 0.9239,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8141333333333334,
      "eval_loss": 0.681545078754425,
      "eval_runtime": 5.4433,
      "eval_samples_per_second": 688.916,
      "eval_steps_per_second": 10.839,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 349.6940612792969,
      "learning_rate": 3.6e-05,
      "loss": 0.5784,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8346666666666667,
      "eval_loss": 0.5405493974685669,
      "eval_runtime": 5.4161,
      "eval_samples_per_second": 692.381,
      "eval_steps_per_second": 10.893,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 90.08492279052734,
      "learning_rate": 1.8e-05,
      "loss": 0.4533,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8474666666666667,
      "eval_loss": 0.49850448966026306,
      "eval_runtime": 5.407,
      "eval_samples_per_second": 693.54,
      "eval_steps_per_second": 10.912,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 146.77029418945312,
      "learning_rate": 0.0,
      "loss": 0.3944,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8533333333333334,
      "eval_loss": 0.47736063599586487,
      "eval_runtime": 5.4778,
      "eval_samples_per_second": 684.588,
      "eval_steps_per_second": 10.771,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}