trainer_state.json

{
  "best_metric": 0.8941333333333333,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0206/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": 38.55060958862305,
      "learning_rate": 0.0003,
      "loss": 1.4831,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8370666666666666,
      "eval_loss": 0.5435140132904053,
      "eval_runtime": 7.1521,
      "eval_samples_per_second": 524.319,
      "eval_steps_per_second": 8.249,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 32.55141067504883,
      "learning_rate": 0.0003,
      "loss": 0.4229,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8672,
      "eval_loss": 0.44658800959587097,
      "eval_runtime": 7.1218,
      "eval_samples_per_second": 526.552,
      "eval_steps_per_second": 8.284,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 20.28790283203125,
      "learning_rate": 0.0003,
      "loss": 0.2636,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8794666666666666,
      "eval_loss": 0.44142499566078186,
      "eval_runtime": 7.0893,
      "eval_samples_per_second": 528.966,
      "eval_steps_per_second": 8.322,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 19.08441925048828,
      "learning_rate": 0.0003,
      "loss": 0.187,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8874666666666666,
      "eval_loss": 0.41485920548439026,
      "eval_runtime": 7.1177,
      "eval_samples_per_second": 526.853,
      "eval_steps_per_second": 8.289,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 24.344465255737305,
      "learning_rate": 0.0003,
      "loss": 0.1397,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8885333333333333,
      "eval_loss": 0.40164339542388916,
      "eval_runtime": 7.1434,
      "eval_samples_per_second": 524.959,
      "eval_steps_per_second": 8.259,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 41.07125473022461,
      "learning_rate": 0.0003,
      "loss": 0.1094,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8941333333333333,
      "eval_loss": 0.4055023193359375,
      "eval_runtime": 7.207,
      "eval_samples_per_second": 520.331,
      "eval_steps_per_second": 8.187,
      "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
}