trainer_state.json

{
  "best_metric": 0.8802666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0775/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": 121.60135650634766,
      "learning_rate": 0.0005,
      "loss": 1.1817,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8541333333333333,
      "eval_loss": 0.48037174344062805,
      "eval_runtime": 13.2014,
      "eval_samples_per_second": 284.061,
      "eval_steps_per_second": 4.469,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 22.47381591796875,
      "learning_rate": 0.0005,
      "loss": 0.3477,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8717333333333334,
      "eval_loss": 0.4240707755088806,
      "eval_runtime": 13.1791,
      "eval_samples_per_second": 284.541,
      "eval_steps_per_second": 4.477,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 23.119586944580078,
      "learning_rate": 0.0005,
      "loss": 0.2185,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4417732357978821,
      "eval_runtime": 13.1626,
      "eval_samples_per_second": 284.898,
      "eval_steps_per_second": 4.482,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 24.113252639770508,
      "learning_rate": 0.0005,
      "loss": 0.1645,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8776,
      "eval_loss": 0.43471816182136536,
      "eval_runtime": 13.5227,
      "eval_samples_per_second": 277.311,
      "eval_steps_per_second": 4.363,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 25.727609634399414,
      "learning_rate": 0.0005,
      "loss": 0.1213,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.45684733986854553,
      "eval_runtime": 13.1965,
      "eval_samples_per_second": 284.166,
      "eval_steps_per_second": 4.471,
      "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
}