trainer_state.json

{
  "best_metric": 0.8826666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0906/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": 32.79265213012695,
      "learning_rate": 0.0003,
      "loss": 1.6604,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8069333333333333,
      "eval_loss": 0.6372787952423096,
      "eval_runtime": 7.443,
      "eval_samples_per_second": 503.831,
      "eval_steps_per_second": 7.927,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 79.12024688720703,
      "learning_rate": 0.0003,
      "loss": 0.563,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8445333333333334,
      "eval_loss": 0.5028535723686218,
      "eval_runtime": 7.4781,
      "eval_samples_per_second": 501.466,
      "eval_steps_per_second": 7.89,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 24.75870704650879,
      "learning_rate": 0.0003,
      "loss": 0.3748,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8562666666666666,
      "eval_loss": 0.4538596272468567,
      "eval_runtime": 7.4679,
      "eval_samples_per_second": 502.151,
      "eval_steps_per_second": 7.901,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 26.89114761352539,
      "learning_rate": 0.0003,
      "loss": 0.2607,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8693333333333333,
      "eval_loss": 0.4444017708301544,
      "eval_runtime": 7.4901,
      "eval_samples_per_second": 500.663,
      "eval_steps_per_second": 7.877,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 36.996463775634766,
      "learning_rate": 0.0003,
      "loss": 0.2085,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8736,
      "eval_loss": 0.427885502576828,
      "eval_runtime": 7.4476,
      "eval_samples_per_second": 503.515,
      "eval_steps_per_second": 7.922,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 33.383541107177734,
      "learning_rate": 0.0003,
      "loss": 0.1529,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8826666666666667,
      "eval_loss": 0.451499879360199,
      "eval_runtime": 7.4764,
      "eval_samples_per_second": 501.578,
      "eval_steps_per_second": 7.891,
      "step": 1998
    }
  ],
  "logging_steps": 500,
  "max_steps": 2331,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "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
}