trainer_state.json

{
  "best_metric": 0.8957333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0371/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": 28.31476402282715,
      "learning_rate": 0.0004809698831278217,
      "loss": 1.4085,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8034666666666667,
      "eval_loss": 0.6844620704650879,
      "eval_runtime": 7.6673,
      "eval_samples_per_second": 489.09,
      "eval_steps_per_second": 7.695,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 18.872425079345703,
      "learning_rate": 0.00042677669529663686,
      "loss": 0.5131,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8496,
      "eval_loss": 0.5051594376564026,
      "eval_runtime": 7.6093,
      "eval_samples_per_second": 492.819,
      "eval_steps_per_second": 7.754,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 61.90553283691406,
      "learning_rate": 0.0003456708580912725,
      "loss": 0.3358,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.4438536465167999,
      "eval_runtime": 7.7013,
      "eval_samples_per_second": 486.928,
      "eval_steps_per_second": 7.661,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 92.3291244506836,
      "learning_rate": 0.00025,
      "loss": 0.2176,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8781333333333333,
      "eval_loss": 0.44612637162208557,
      "eval_runtime": 7.6122,
      "eval_samples_per_second": 492.628,
      "eval_steps_per_second": 7.751,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 30.883535385131836,
      "learning_rate": 0.00015432914190872756,
      "loss": 0.1358,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8872,
      "eval_loss": 0.39638057351112366,
      "eval_runtime": 7.6256,
      "eval_samples_per_second": 491.763,
      "eval_steps_per_second": 7.737,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 18.522443771362305,
      "learning_rate": 7.322330470336314e-05,
      "loss": 0.0779,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8922666666666667,
      "eval_loss": 0.4251933693885803,
      "eval_runtime": 7.7476,
      "eval_samples_per_second": 484.023,
      "eval_steps_per_second": 7.615,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 23.255895614624023,
      "learning_rate": 1.9030116872178316e-05,
      "loss": 0.0527,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8957333333333334,
      "eval_loss": 0.42178595066070557,
      "eval_runtime": 7.6278,
      "eval_samples_per_second": 491.622,
      "eval_steps_per_second": 7.735,
      "step": 2331
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}