trainer_state.json

{
  "best_metric": 0.8744,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0782/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 56.736083984375,
      "learning_rate": 4.752422169756048e-05,
      "loss": 3.6859,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5272,
      "eval_loss": 3.191502094268799,
      "eval_runtime": 14.3505,
      "eval_samples_per_second": 261.314,
      "eval_steps_per_second": 4.111,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 76.31793212890625,
      "learning_rate": 4.058724504646834e-05,
      "loss": 1.6838,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7826666666666666,
      "eval_loss": 0.8968120813369751,
      "eval_runtime": 14.2465,
      "eval_samples_per_second": 263.222,
      "eval_steps_per_second": 4.141,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 125.69475555419922,
      "learning_rate": 3.056302334890786e-05,
      "loss": 0.7174,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8413333333333334,
      "eval_loss": 0.582084059715271,
      "eval_runtime": 14.1611,
      "eval_samples_per_second": 264.811,
      "eval_steps_per_second": 4.166,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 129.65609741210938,
      "learning_rate": 1.9436976651092144e-05,
      "loss": 0.496,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.501431941986084,
      "eval_runtime": 14.1821,
      "eval_samples_per_second": 264.417,
      "eval_steps_per_second": 4.16,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 223.1921844482422,
      "learning_rate": 9.412754953531663e-06,
      "loss": 0.4076,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8674666666666667,
      "eval_loss": 0.4709937572479248,
      "eval_runtime": 14.2019,
      "eval_samples_per_second": 264.049,
      "eval_steps_per_second": 4.154,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 235.7877197265625,
      "learning_rate": 2.475778302439524e-06,
      "loss": 0.3654,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8744,
      "eval_loss": 0.4566883444786072,
      "eval_runtime": 14.0733,
      "eval_samples_per_second": 266.462,
      "eval_steps_per_second": 4.192,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}