trainer_state.json

{
  "best_metric": 0.8504,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0711/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 149.3629913330078,
      "learning_rate": 4.752422169756048e-05,
      "loss": 3.6023,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.568,
      "eval_loss": 2.8812625408172607,
      "eval_runtime": 7.0382,
      "eval_samples_per_second": 532.805,
      "eval_steps_per_second": 8.383,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 212.52784729003906,
      "learning_rate": 4.058724504646834e-05,
      "loss": 1.6093,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7570666666666667,
      "eval_loss": 0.8762533664703369,
      "eval_runtime": 6.943,
      "eval_samples_per_second": 540.113,
      "eval_steps_per_second": 8.498,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 266.45343017578125,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.7284,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8237333333333333,
      "eval_loss": 0.6275570392608643,
      "eval_runtime": 6.9714,
      "eval_samples_per_second": 537.915,
      "eval_steps_per_second": 8.463,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 162.6540069580078,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.5277,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8314666666666667,
      "eval_loss": 0.563921332359314,
      "eval_runtime": 7.6838,
      "eval_samples_per_second": 488.039,
      "eval_steps_per_second": 7.678,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 360.763916015625,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.4359,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8488,
      "eval_loss": 0.5265204906463623,
      "eval_runtime": 7.0163,
      "eval_samples_per_second": 534.472,
      "eval_steps_per_second": 8.409,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 291.88763427734375,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.3912,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8370666666666666,
      "eval_loss": 0.5404622554779053,
      "eval_runtime": 6.9796,
      "eval_samples_per_second": 537.278,
      "eval_steps_per_second": 8.453,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 104.15003967285156,
      "learning_rate": 0.0,
      "loss": 0.3757,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8504,
      "eval_loss": 0.5015265345573425,
      "eval_runtime": 6.9705,
      "eval_samples_per_second": 537.979,
      "eval_steps_per_second": 8.464,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}