trainer_state.json

{
  "best_metric": 0.7904,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0496/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": 81.09800720214844,
      "learning_rate": 4e-05,
      "loss": 3.6945,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4922666666666667,
      "eval_loss": 3.261120319366455,
      "eval_runtime": 13.2013,
      "eval_samples_per_second": 284.064,
      "eval_steps_per_second": 4.469,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 111.64807891845703,
      "learning_rate": 3e-05,
      "loss": 2.196,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6909333333333333,
      "eval_loss": 1.3831665515899658,
      "eval_runtime": 13.1366,
      "eval_samples_per_second": 285.462,
      "eval_steps_per_second": 4.491,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 134.1149444580078,
      "learning_rate": 2e-05,
      "loss": 1.1291,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7653333333333333,
      "eval_loss": 0.9066988229751587,
      "eval_runtime": 13.2618,
      "eval_samples_per_second": 282.768,
      "eval_steps_per_second": 4.449,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 251.1747283935547,
      "learning_rate": 1e-05,
      "loss": 0.8634,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7874666666666666,
      "eval_loss": 0.7657472491264343,
      "eval_runtime": 13.0765,
      "eval_samples_per_second": 286.774,
      "eval_steps_per_second": 4.512,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 124.97953033447266,
      "learning_rate": 0.0,
      "loss": 0.7831,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7904,
      "eval_loss": 0.7454047799110413,
      "eval_runtime": 13.0845,
      "eval_samples_per_second": 286.599,
      "eval_steps_per_second": 4.509,
      "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
}