trainer_state.json

{
  "best_metric": 0.8784,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0643/checkpoints/checkpoint-2997",
  "epoch": 9.0,
  "eval_steps": 500,
  "global_step": 2997,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 55.71562194824219,
      "learning_rate": 7e-05,
      "loss": 3.3802,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5824,
      "eval_loss": 1.8887196779251099,
      "eval_runtime": 8.0167,
      "eval_samples_per_second": 467.775,
      "eval_steps_per_second": 7.36,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 246.6086883544922,
      "learning_rate": 7e-05,
      "loss": 1.1212,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8005333333333333,
      "eval_loss": 0.7105347514152527,
      "eval_runtime": 7.4747,
      "eval_samples_per_second": 501.692,
      "eval_steps_per_second": 7.893,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 96.97613525390625,
      "learning_rate": 7e-05,
      "loss": 0.6549,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.836,
      "eval_loss": 0.5566152334213257,
      "eval_runtime": 7.881,
      "eval_samples_per_second": 475.831,
      "eval_steps_per_second": 7.486,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 82.99352264404297,
      "learning_rate": 7e-05,
      "loss": 0.4951,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8610666666666666,
      "eval_loss": 0.4799044728279114,
      "eval_runtime": 7.4729,
      "eval_samples_per_second": 501.811,
      "eval_steps_per_second": 7.895,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 199.56512451171875,
      "learning_rate": 7e-05,
      "loss": 0.3919,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8616,
      "eval_loss": 0.47139081358909607,
      "eval_runtime": 7.4481,
      "eval_samples_per_second": 503.486,
      "eval_steps_per_second": 7.922,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 199.5041046142578,
      "learning_rate": 7e-05,
      "loss": 0.3317,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.43580183386802673,
      "eval_runtime": 7.4923,
      "eval_samples_per_second": 500.512,
      "eval_steps_per_second": 7.875,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 66.11186218261719,
      "learning_rate": 7e-05,
      "loss": 0.2676,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8754666666666666,
      "eval_loss": 0.4195924997329712,
      "eval_runtime": 7.4561,
      "eval_samples_per_second": 502.945,
      "eval_steps_per_second": 7.913,
      "step": 2331
    },
    {
      "epoch": 8.0,
      "grad_norm": 49.9759521484375,
      "learning_rate": 7e-05,
      "loss": 0.2232,
      "step": 2664
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8778666666666667,
      "eval_loss": 0.40542715787887573,
      "eval_runtime": 7.42,
      "eval_samples_per_second": 505.393,
      "eval_steps_per_second": 7.952,
      "step": 2664
    },
    {
      "epoch": 9.0,
      "grad_norm": 74.59369659423828,
      "learning_rate": 7e-05,
      "loss": 0.1886,
      "step": 2997
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8784,
      "eval_loss": 0.4242912232875824,
      "eval_runtime": 7.6797,
      "eval_samples_per_second": 488.298,
      "eval_steps_per_second": 7.683,
      "step": 2997
    }
  ],
  "logging_steps": 500,
  "max_steps": 2997,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "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": 7.3587863155968e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}