trainer_state.json

{
  "best_metric": 0.8562666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0257/checkpoints/checkpoint-2664",
  "epoch": 8.0,
  "eval_steps": 500,
  "global_step": 2664,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 169.8601531982422,
      "learning_rate": 4.8096988312782174e-05,
      "loss": 3.5767,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.568,
      "eval_loss": 2.8798787593841553,
      "eval_runtime": 7.4991,
      "eval_samples_per_second": 500.06,
      "eval_steps_per_second": 7.868,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 174.6171112060547,
      "learning_rate": 4.267766952966369e-05,
      "loss": 1.6205,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7704,
      "eval_loss": 0.8748669028282166,
      "eval_runtime": 8.1937,
      "eval_samples_per_second": 457.666,
      "eval_steps_per_second": 7.201,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 89.88113403320312,
      "learning_rate": 3.456708580912725e-05,
      "loss": 0.7241,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.82,
      "eval_loss": 0.6316601037979126,
      "eval_runtime": 7.7385,
      "eval_samples_per_second": 484.588,
      "eval_steps_per_second": 7.624,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 230.87887573242188,
      "learning_rate": 2.5e-05,
      "loss": 0.5221,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8397333333333333,
      "eval_loss": 0.5298848152160645,
      "eval_runtime": 7.577,
      "eval_samples_per_second": 494.917,
      "eval_steps_per_second": 7.787,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 283.05377197265625,
      "learning_rate": 1.5432914190872757e-05,
      "loss": 0.4232,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8525333333333334,
      "eval_loss": 0.4884917438030243,
      "eval_runtime": 7.6547,
      "eval_samples_per_second": 489.897,
      "eval_steps_per_second": 7.708,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 162.71090698242188,
      "learning_rate": 7.3223304703363135e-06,
      "loss": 0.3682,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8549333333333333,
      "eval_loss": 0.4718301296234131,
      "eval_runtime": 7.8724,
      "eval_samples_per_second": 476.349,
      "eval_steps_per_second": 7.495,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 360.67578125,
      "learning_rate": 1.9030116872178316e-06,
      "loss": 0.3447,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.856,
      "eval_loss": 0.4570256173610687,
      "eval_runtime": 7.8744,
      "eval_samples_per_second": 476.228,
      "eval_steps_per_second": 7.493,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 417.260986328125,
      "learning_rate": 0.0,
      "loss": 0.3348,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.4626965820789337,
      "eval_runtime": 7.6572,
      "eval_samples_per_second": 489.738,
      "eval_steps_per_second": 7.705,
      "step": 2664
    }
  ],
  "logging_steps": 500,
  "max_steps": 2664,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "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": 6.5411433916416e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}