trainer_state.json

{
  "best_metric": 0.9,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0252/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 140.84600830078125,
      "learning_rate": 5e-05,
      "loss": 3.5992,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5309333333333334,
      "eval_loss": 2.8154983520507812,
      "eval_runtime": 8.3233,
      "eval_samples_per_second": 450.541,
      "eval_steps_per_second": 7.089,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 119.87128448486328,
      "learning_rate": 5e-05,
      "loss": 1.5109,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7954666666666667,
      "eval_loss": 0.7709070444107056,
      "eval_runtime": 8.0862,
      "eval_samples_per_second": 463.751,
      "eval_steps_per_second": 7.296,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 37.907371520996094,
      "learning_rate": 5e-05,
      "loss": 0.6133,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8464,
      "eval_loss": 0.50244140625,
      "eval_runtime": 8.0238,
      "eval_samples_per_second": 467.36,
      "eval_steps_per_second": 7.353,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 104.55272674560547,
      "learning_rate": 5e-05,
      "loss": 0.4233,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8709333333333333,
      "eval_loss": 0.41710352897644043,
      "eval_runtime": 8.036,
      "eval_samples_per_second": 466.649,
      "eval_steps_per_second": 7.342,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 199.27528381347656,
      "learning_rate": 5e-05,
      "loss": 0.3206,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8850666666666667,
      "eval_loss": 0.37746718525886536,
      "eval_runtime": 8.082,
      "eval_samples_per_second": 463.993,
      "eval_steps_per_second": 7.3,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 134.68946838378906,
      "learning_rate": 5e-05,
      "loss": 0.2486,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8904,
      "eval_loss": 0.35554370284080505,
      "eval_runtime": 8.0692,
      "eval_samples_per_second": 464.729,
      "eval_steps_per_second": 7.312,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 75.631103515625,
      "learning_rate": 5e-05,
      "loss": 0.1972,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8861333333333333,
      "eval_loss": 0.364734947681427,
      "eval_runtime": 8.0695,
      "eval_samples_per_second": 464.712,
      "eval_steps_per_second": 7.311,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 105.85884094238281,
      "learning_rate": 5e-05,
      "loss": 0.1619,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9,
      "eval_loss": 0.31833526492118835,
      "eval_runtime": 8.0625,
      "eval_samples_per_second": 465.117,
      "eval_steps_per_second": 7.318,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}