trainer_state.json

{
  "best_metric": 0.8101333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0004/checkpoints/checkpoint-1665",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 1665,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 244.82933044433594,
      "learning_rate": 4.522542485937369e-05,
      "loss": 3.7001,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4845333333333333,
      "eval_loss": 3.146526336669922,
      "eval_runtime": 8.8852,
      "eval_samples_per_second": 422.048,
      "eval_steps_per_second": 6.64,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 221.75074768066406,
      "learning_rate": 3.272542485937369e-05,
      "loss": 2.0873,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7210666666666666,
      "eval_loss": 1.2446309328079224,
      "eval_runtime": 9.0091,
      "eval_samples_per_second": 416.244,
      "eval_steps_per_second": 6.549,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 149.58526611328125,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 1.0481,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7845333333333333,
      "eval_loss": 0.8526014685630798,
      "eval_runtime": 9.4011,
      "eval_samples_per_second": 398.891,
      "eval_steps_per_second": 6.276,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 248.8019561767578,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.8101,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8064,
      "eval_loss": 0.7515445947647095,
      "eval_runtime": 8.9694,
      "eval_samples_per_second": 418.088,
      "eval_steps_per_second": 6.578,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 331.9724426269531,
      "learning_rate": 0.0,
      "loss": 0.7673,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8101333333333334,
      "eval_loss": 0.7245769500732422,
      "eval_runtime": 9.0175,
      "eval_samples_per_second": 415.86,
      "eval_steps_per_second": 6.543,
      "step": 1665
    }
  ],
  "logging_steps": 500,
  "max_steps": 1665,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 4.088214619776e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}