trainer_state.json

{
  "best_metric": 0.8853333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0214/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": 39.149017333984375,
      "learning_rate": 0.0003,
      "loss": 1.6554,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8093333333333333,
      "eval_loss": 0.6147518157958984,
      "eval_runtime": 13.417,
      "eval_samples_per_second": 279.497,
      "eval_steps_per_second": 4.397,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 38.94075012207031,
      "learning_rate": 0.0003,
      "loss": 0.5219,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.4630081355571747,
      "eval_runtime": 13.4418,
      "eval_samples_per_second": 278.981,
      "eval_steps_per_second": 4.389,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 35.62356185913086,
      "learning_rate": 0.0003,
      "loss": 0.3559,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8645333333333334,
      "eval_loss": 0.4533454477787018,
      "eval_runtime": 13.4806,
      "eval_samples_per_second": 278.178,
      "eval_steps_per_second": 4.377,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 51.48064422607422,
      "learning_rate": 0.0003,
      "loss": 0.2534,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8725333333333334,
      "eval_loss": 0.4493444561958313,
      "eval_runtime": 13.3302,
      "eval_samples_per_second": 281.317,
      "eval_steps_per_second": 4.426,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 60.846397399902344,
      "learning_rate": 0.0003,
      "loss": 0.2009,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.868,
      "eval_loss": 0.46974751353263855,
      "eval_runtime": 13.657,
      "eval_samples_per_second": 274.585,
      "eval_steps_per_second": 4.32,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 24.87139892578125,
      "learning_rate": 0.0003,
      "loss": 0.1448,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8821333333333333,
      "eval_loss": 0.44294655323028564,
      "eval_runtime": 13.2845,
      "eval_samples_per_second": 282.284,
      "eval_steps_per_second": 4.441,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 31.59549331665039,
      "learning_rate": 0.0003,
      "loss": 0.1247,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8853333333333333,
      "eval_loss": 0.44069886207580566,
      "eval_runtime": 13.9888,
      "eval_samples_per_second": 268.072,
      "eval_steps_per_second": 4.218,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}