trainer_state.json

{
  "best_metric": 0.8746666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0796/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": 138.9960479736328,
      "learning_rate": 0.0001,
      "loss": 2.773,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7045333333333333,
      "eval_loss": 1.0935144424438477,
      "eval_runtime": 7.3902,
      "eval_samples_per_second": 507.426,
      "eval_steps_per_second": 7.984,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 100.66458892822266,
      "learning_rate": 0.0001,
      "loss": 0.7427,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8221333333333334,
      "eval_loss": 0.5656073093414307,
      "eval_runtime": 7.3701,
      "eval_samples_per_second": 508.814,
      "eval_steps_per_second": 8.005,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 136.41845703125,
      "learning_rate": 0.0001,
      "loss": 0.4613,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8504,
      "eval_loss": 0.4662906229496002,
      "eval_runtime": 7.3539,
      "eval_samples_per_second": 509.934,
      "eval_steps_per_second": 8.023,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 95.06509399414062,
      "learning_rate": 0.0001,
      "loss": 0.3312,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.42493242025375366,
      "eval_runtime": 10.948,
      "eval_samples_per_second": 342.528,
      "eval_steps_per_second": 5.389,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 144.40948486328125,
      "learning_rate": 0.0001,
      "loss": 0.252,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8722666666666666,
      "eval_loss": 0.41476041078567505,
      "eval_runtime": 7.3587,
      "eval_samples_per_second": 509.598,
      "eval_steps_per_second": 8.018,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 81.42378234863281,
      "learning_rate": 0.0001,
      "loss": 0.1919,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8690666666666667,
      "eval_loss": 0.4343946874141693,
      "eval_runtime": 7.3559,
      "eval_samples_per_second": 509.798,
      "eval_steps_per_second": 8.021,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 48.117008209228516,
      "learning_rate": 0.0001,
      "loss": 0.1558,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8746666666666667,
      "eval_loss": 0.41153937578201294,
      "eval_runtime": 7.5242,
      "eval_samples_per_second": 498.389,
      "eval_steps_per_second": 7.841,
      "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
}