trainer_state.json

{
  "best_metric": 0.7909333333333334,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0359/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": 122.93004608154297,
      "learning_rate": 4.522542485937369e-05,
      "loss": 3.6945,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.49466666666666664,
      "eval_loss": 3.1414966583251953,
      "eval_runtime": 13.0528,
      "eval_samples_per_second": 287.294,
      "eval_steps_per_second": 4.52,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 99.42825317382812,
      "learning_rate": 3.272542485937369e-05,
      "loss": 2.0824,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6984,
      "eval_loss": 1.2925070524215698,
      "eval_runtime": 12.9309,
      "eval_samples_per_second": 290.002,
      "eval_steps_per_second": 4.563,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 151.95187377929688,
      "learning_rate": 1.7274575140626318e-05,
      "loss": 1.1211,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7666666666666667,
      "eval_loss": 0.8968787789344788,
      "eval_runtime": 12.9323,
      "eval_samples_per_second": 289.971,
      "eval_steps_per_second": 4.562,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 343.5714416503906,
      "learning_rate": 4.7745751406263165e-06,
      "loss": 0.8941,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7848,
      "eval_loss": 0.8145498633384705,
      "eval_runtime": 12.9637,
      "eval_samples_per_second": 289.27,
      "eval_steps_per_second": 4.551,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 166.5993194580078,
      "learning_rate": 0.0,
      "loss": 0.8253,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7909333333333334,
      "eval_loss": 0.8153493404388428,
      "eval_runtime": 12.9685,
      "eval_samples_per_second": 289.162,
      "eval_steps_per_second": 4.549,
      "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
}