trainer_state.json

{
  "best_metric": 0.8802666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0689/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": 162.62161254882812,
      "learning_rate": 8.75e-05,
      "loss": 2.7267,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7317333333333333,
      "eval_loss": 0.9681093096733093,
      "eval_runtime": 5.4545,
      "eval_samples_per_second": 687.506,
      "eval_steps_per_second": 10.817,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 142.60888671875,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.7203,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8314666666666667,
      "eval_loss": 0.5368010997772217,
      "eval_runtime": 5.0987,
      "eval_samples_per_second": 735.483,
      "eval_steps_per_second": 11.572,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 132.53778076171875,
      "learning_rate": 6.25e-05,
      "loss": 0.4471,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8586666666666667,
      "eval_loss": 0.43269649147987366,
      "eval_runtime": 5.0866,
      "eval_samples_per_second": 737.238,
      "eval_steps_per_second": 11.599,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 63.15406036376953,
      "learning_rate": 5e-05,
      "loss": 0.3303,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8698666666666667,
      "eval_loss": 0.41298720240592957,
      "eval_runtime": 5.0715,
      "eval_samples_per_second": 739.422,
      "eval_steps_per_second": 11.634,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 242.4856719970703,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.2561,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8770666666666667,
      "eval_loss": 0.3835950493812561,
      "eval_runtime": 5.0723,
      "eval_samples_per_second": 739.307,
      "eval_steps_per_second": 11.632,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 68.03028869628906,
      "learning_rate": 2.5e-05,
      "loss": 0.2056,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.3760527968406677,
      "eval_runtime": 5.1009,
      "eval_samples_per_second": 735.158,
      "eval_steps_per_second": 11.566,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 197.464599609375,
      "learning_rate": 1.25e-05,
      "loss": 0.1794,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.3842315375804901,
      "eval_runtime": 5.0539,
      "eval_samples_per_second": 742.003,
      "eval_steps_per_second": 11.674,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 69.98363494873047,
      "learning_rate": 0.0,
      "loss": 0.1562,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8802666666666666,
      "eval_loss": 0.3783366084098816,
      "eval_runtime": 5.0686,
      "eval_samples_per_second": 739.85,
      "eval_steps_per_second": 11.64,
      "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
}