trainer_state.json

{
  "best_metric": 0.9002666666666667,
  "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0354/checkpoints/checkpoint-2331",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 2331,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 44.14098358154297,
      "learning_rate": 0.0003,
      "loss": 1.443,
      "step": 333
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8581333333333333,
      "eval_loss": 0.4630856215953827,
      "eval_runtime": 14.6543,
      "eval_samples_per_second": 255.897,
      "eval_steps_per_second": 4.026,
      "step": 333
    },
    {
      "epoch": 2.0,
      "grad_norm": 27.342559814453125,
      "learning_rate": 0.0003,
      "loss": 0.3904,
      "step": 666
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8786666666666667,
      "eval_loss": 0.4165979027748108,
      "eval_runtime": 14.5407,
      "eval_samples_per_second": 257.897,
      "eval_steps_per_second": 4.058,
      "step": 666
    },
    {
      "epoch": 3.0,
      "grad_norm": 54.09717559814453,
      "learning_rate": 0.0003,
      "loss": 0.2419,
      "step": 999
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8946666666666667,
      "eval_loss": 0.36788827180862427,
      "eval_runtime": 14.5309,
      "eval_samples_per_second": 258.071,
      "eval_steps_per_second": 4.06,
      "step": 999
    },
    {
      "epoch": 4.0,
      "grad_norm": 21.16099739074707,
      "learning_rate": 0.0003,
      "loss": 0.1607,
      "step": 1332
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8949333333333334,
      "eval_loss": 0.38946816325187683,
      "eval_runtime": 14.6049,
      "eval_samples_per_second": 256.763,
      "eval_steps_per_second": 4.04,
      "step": 1332
    },
    {
      "epoch": 5.0,
      "grad_norm": 28.864116668701172,
      "learning_rate": 0.0003,
      "loss": 0.1215,
      "step": 1665
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8965333333333333,
      "eval_loss": 0.39862462878227234,
      "eval_runtime": 14.4715,
      "eval_samples_per_second": 259.13,
      "eval_steps_per_second": 4.077,
      "step": 1665
    },
    {
      "epoch": 6.0,
      "grad_norm": 42.43585205078125,
      "learning_rate": 0.0003,
      "loss": 0.1015,
      "step": 1998
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8992,
      "eval_loss": 0.407478004693985,
      "eval_runtime": 14.414,
      "eval_samples_per_second": 260.163,
      "eval_steps_per_second": 4.093,
      "step": 1998
    },
    {
      "epoch": 7.0,
      "grad_norm": 40.835693359375,
      "learning_rate": 0.0003,
      "loss": 0.0745,
      "step": 2331
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9002666666666667,
      "eval_loss": 0.41155481338500977,
      "eval_runtime": 14.6158,
      "eval_samples_per_second": 256.571,
      "eval_steps_per_second": 4.037,
      "step": 2331
    }
  ],
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 5.7235004676864e+18,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}