trainer_state.json

{
  "best_metric": 0.8874666666666666,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0464/checkpoints/checkpoint-1998",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 1998,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 98.41361999511719,
      "learning_rate": 0.0003,
      "loss": 1.5019,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8402666666666667,
      "eval_loss": 0.5347887277603149,
      "eval_runtime": 7.0342,
      "eval_samples_per_second": 533.107,
      "eval_steps_per_second": 8.388,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 29.615398406982422,
      "learning_rate": 0.0003,
      "loss": 0.3912,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8573333333333333,
      "eval_loss": 0.4815865457057953,
      "eval_runtime": 6.9445,
      "eval_samples_per_second": 539.998,
      "eval_steps_per_second": 8.496,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 29.83663558959961,
      "learning_rate": 0.0003,
      "loss": 0.2253,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8762666666666666,
      "eval_loss": 0.4207737445831299,
      "eval_runtime": 6.9518,
      "eval_samples_per_second": 539.43,
      "eval_steps_per_second": 8.487,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 39.87559127807617,
      "learning_rate": 0.0003,
      "loss": 0.1506,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8797333333333334,
      "eval_loss": 0.4406000077724457,
      "eval_runtime": 6.9438,
      "eval_samples_per_second": 540.053,
      "eval_steps_per_second": 8.497,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 37.399513244628906,
      "learning_rate": 0.0003,
      "loss": 0.0989,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8741333333333333,
      "eval_loss": 0.5118945837020874,
      "eval_runtime": 6.9556,
      "eval_samples_per_second": 539.131,
      "eval_steps_per_second": 8.482,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 48.573951721191406,
      "learning_rate": 0.0003,
      "loss": 0.0924,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.4842020869255066,
      "eval_runtime": 7.104,
      "eval_samples_per_second": 527.868,
      "eval_steps_per_second": 8.305,
      "step": 1998
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.9058575437312e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}