trainer_state.json

{
  "best_metric": 0.8813333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0484/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": 136.24862670898438,
      "learning_rate": 7e-05,
      "loss": 3.3695,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6224,
      "eval_loss": 1.776260256767273,
      "eval_runtime": 15.4557,
      "eval_samples_per_second": 242.629,
      "eval_steps_per_second": 3.817,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 293.82452392578125,
      "learning_rate": 7e-05,
      "loss": 1.0214,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7922666666666667,
      "eval_loss": 0.6821350455284119,
      "eval_runtime": 15.175,
      "eval_samples_per_second": 247.117,
      "eval_steps_per_second": 3.888,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 79.00701141357422,
      "learning_rate": 7e-05,
      "loss": 0.5698,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8493333333333334,
      "eval_loss": 0.4915453791618347,
      "eval_runtime": 15.2169,
      "eval_samples_per_second": 246.437,
      "eval_steps_per_second": 3.877,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 114.53131103515625,
      "learning_rate": 7e-05,
      "loss": 0.411,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8677333333333334,
      "eval_loss": 0.43471333384513855,
      "eval_runtime": 15.0528,
      "eval_samples_per_second": 249.123,
      "eval_steps_per_second": 3.92,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 106.54593658447266,
      "learning_rate": 7e-05,
      "loss": 0.313,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8813333333333333,
      "eval_loss": 0.4056074917316437,
      "eval_runtime": 15.0768,
      "eval_samples_per_second": 248.726,
      "eval_steps_per_second": 3.913,
      "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
}